System

Abstract

A system is provided.SOLUTION: A system comprising: means for indexing necessary information from a large amount of information; means for receiving a search keyword from a user; means for extracting a relevant document from the indexed data based on the search keyword; means for displaying a search result to the user; and means for recording a search history of the user and learning a search tendency.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In today's world, where a vast amount of information is available, it is difficult to quickly and accurately search for necessary information, which leads to a decline in business efficiency and sluggish productivity. Conventional information management tools and search engines lack the functionality to learn users' search trends and provide optimal search results, making it difficult to provide search results that are appropriate for each individual user. The objective of the present invention is to solve this problem and improve the information search experience for each individual user. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for indexing necessary information from a large amount of information, a means for receiving search keywords from a user, a means for extracting related documents from the indexed data based on the search keywords, a means for displaying search results to the user, and a means for recording the user's search history and learning search trends. The system also achieves efficient information search by combining a means for updating a machine learning model based on a search query input to reflect the user's trends in the next search, and a means for using a generative AI model to quickly and accurately provide information related to the user's search keywords. This significantly reduces the time required for information search, improving business efficiency and productivity.

[0006] "Large amounts of information" refers to the wide variety of data stored in documents and databases used by users.

[0007] "Indexing" refers to the process of converting data into a specific structure and creating an index to speed up searches.

[0008] "Search keywords" refer to words or phrases that users enter into their devices when searching for specific information.

[0009] "Relevant Documents" refers to documents or data that match a search query and are extracted from indexed data based on search keywords.

[0010] "Search history" refers to a record of past search queries and search results viewed by a user.

[0011] "Search trends" refers to the direction of a user's preferences and interests, which are inferred based on the user's search history and behavioral patterns.

[0012] A "machine learning model" refers to an algorithm that uses computational methods to make predictions and classifications based on data.

[0013] A "generative AI model" refers to an algorithm that uses artificial intelligence technology to understand the context of input data and generate optimal search results. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0022] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0035] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends. This system works in cooperation with a server, a terminal, and a user.

[0036] Data Indexing

[0037] The server first retrieves all document information from the content management system and indexes them. The indexed data is used to speed up searches. Indexing occurs automatically whenever a new document is added.

[0038] Enter search keywords

[0039] A user inputs search keywords into a terminal to search for specific information, and the search keywords are transmitted to a server via the terminal.

[0040] Find related documents

[0041] The server searches the index based on the received search keywords to extract the most relevant documents, using rapid references to the indexed data to generate optimal search results.

[0042] Displaying search results

[0043] The terminal displays the search results received from the server to the user. The displayed search results include titles and paths, which the user can refer to to access the desired information.

[0044] Record and learn from your search history

[0045] The server records the user's search history and updates the machine learning model based on this information, allowing the user's search trends to be reflected in the next search, resulting in more relevant search results.

[0046] Using generative AI models

[0047] Additionally, the server uses a generative AI model to understand the context of the search keywords entered and generate optimal search results, which goes beyond simple keyword matching to achieve more advanced searches.

[0048] Specific examples

[0049] For example, consider a case where a user searches for information about "project management." The user types "project management" into a terminal. The terminal sends this input to a server. The server searches the index based on the keyword "project management" and extracts relevant documents.

[0050] As a search result, the server extracts documents with titles such as "Project Management Fundamentals" and "Efficient Project Management Methods" and sends them to the terminal. The terminal displays them to the user, allowing the user to quickly access the information they need.

[0051] The server also records the user's search history and learns to prioritize displaying information related to "project management" in subsequent searches.

[0052] By combining users' search history with generative AI models, this system can provide more accurate and personalized search results, significantly reducing the time required to search for information.

[0053] In this way, Knowledge Comet provides users with a fast and accurate information search experience, contributing to improved overall business efficiency and productivity.

[0054] The processing flow will be explained below.

[0055] Step 1:

[0056] The server retrieves all document information from the content management system, indexes these documents, transforms the indexed data into a structure that facilitates fast searches, and stores it in an index data store.

[0057] Step 2:

[0058] A user inputs search keywords into a terminal to search for specific information. The search keywords input by a user are specific words or phrases.

[0059] Step 3:

[0060] The terminal receives search keywords entered by the user and transmits this information to the server, where the search keywords are processed.

[0061] Step 4:

[0062] The server takes the received search keywords and searches for relevant documents from the index data store. The search query is analyzed by a parser to extract the most relevant documents.

[0063] Step 5:

[0064] The server compiles the extracted related document information (e.g., title, path) as search results, and formats the search results in a user-friendly format.

[0065] Step 6:

[0066] The server sends the formatted search results to the terminal, which include the document title and path.

[0067] Step 7:

[0068] The terminal displays the search results received from the server to the user, who can then refer to the displayed results and access the information they need.

[0069] Step 8:

[0070] The server records the user's search queries and stores them as the user's search history, which is associated with the user's profile.

[0071] Step 9:

[0072] The server uses the saved search history to update the machine learning model, which is then used to provide better search results the next time the user searches.

[0073] Step 10:

[0074] The server uses a generative AI model to understand the context of the user's search keywords, generating advanced search results that go beyond simple keyword matching.

[0075] Step 11:

[0076] Users will be able to get more relevant search results faster on subsequent searches, improving their information-seeking experience.

[0077] Example 1

[0078] 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."

[0079] Conventional information retrieval systems have difficulty quickly and accurately searching for the necessary information from a large amount of information, and it is difficult to provide personalized search results that appropriately reflect the user's search trends. Furthermore, because they rely on simple keyword matching, they are unable to realize advanced searches that understand the context. Furthermore, when a user conducts multiple searches, the system lacks the ability to fully learn the user's search history and utilize it for future searches.

[0080] 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.

[0081] In this invention, the server includes: means for indexing necessary information from a large amount of information; means for receiving search keywords from a user; means for extracting related data from the indexed data based on the search keywords; means for displaying search results to the user; means for recording the user's search history and learning search trends; and means for analyzing the context of the entered search keywords using a generative AI model to generate optimal search results. By combining the user's search history and the generative AI model, it is possible to provide more accurate and personalized search results and significantly reduce the time required for information search.

[0082] "Indexing" is the process of organizing and classifying large amounts of information based on specific criteria so that it can be easily searched.

[0083] A "search keyword" is a word or phrase that a user enters when searching for specific information.

[0084] "Related data" refers to data items related to the targeted information that are extracted based on the user's search keywords.

[0085] "Search results" are a list of related data returned in response to a user's search keywords and displayed to the user.

[0086] "Search history" is a record of searches that a user has performed up to now, and includes information such as date and time and keywords.

[0087] "Search trends" are patterns and characteristics extracted from a user's past search behavior.

[0088] A "generative AI model" is a mathematical model that uses artificial intelligence to analyze the context and meaning of input data and generate appropriate output.

[0089] "Context" is other information or data that surrounds a keyword or statement and influences its meaning and interpretation.

[0090] "Personalization" means providing services and content that are specifically tailored and optimized based on individual user characteristics and behavior.

[0091] "Information retrieval" is the process of extracting specific information from a particular database or index.

[0092] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends. This system works in cooperation with a server, a terminal, and a user.

[0093] Data Indexing

[0094] The server first retrieves all document information from the content management system and indexes it. This process uses search engine software such as ElasticSearch or Apache Lucene. Every time a new document is added, the server automatically updates the index, speeding up searches.

[0095] Enter search keywords

[0096] A user inputs keywords into the device to search for specific information. For example, if a user inputs the keyword "project management," the device sends this input to the server via a REST API. The search keywords are sent to the server in JSON format using an AJAX request.

[0097] Find related documents

[0098] The server searches the index based on the received search keywords, generates and executes an ElasticSearch search query to extract the most relevant documents, and then packages the extracted document metadata (title, path, etc.) in JSON format and sends it to the device as a response.

[0099] Displaying search results

[0100] The terminal displays the search results received from the server to the user. Specifically, JavaScript is used to dynamically generate HTML elements and create a search result list. For example, the search results may display titles such as "Project Management Basics" and "Efficient Project Management Methods," with each title serving as a link to the corresponding document's detailed page.

[0101] Record and learn from your search history

[0102] The server records the user's search history. Specifically, it stores the search keywords and search execution timestamps in MongoDB, a NoSQL database. This recorded data is used to update machine learning models (e.g., TensorFlow and Scikit-learn) and learn the user's search patterns. This allows the server to provide more personalized search results the next time the user searches.

[0103] Using generative AI models

[0104] Furthermore, the server uses a generative AI model to understand the context of the search keywords entered. For example, by using OpenAI's GPT-3 model, it achieves a more advanced search than simple keyword matching. The generative AI model understands the context and generates advanced search results. Specific prompts could be something like:

[0105] "What are some basic project management information?"

[0106] "I'd like to get an overview of efficient project management methods."

[0107] Specific examples

[0108] For example, a user may search for information related to "project management." The user types "project management" into their device, which then sends this input to the server via an AJAX request. The server uses ElasticSearch to extract relevant documents from the index and returns the document information to the device. The device displays the search results to the user, who can click links to access more information. The server then records this search history and learns to prioritize displaying information related to "project management" in future searches.

[0109] By combining users' search history with generative AI models, the system can provide more accurate and personalized search results, significantly reducing the time it takes to find information.

[0110] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0111] Step 1: Index the data

[0112] The server first retrieves all document information from the content management system. This is done using SQL queries or API requests. To index the retrieved document information, ElasticSearch is used to organize and classify the data and create an index that enables fast searches. This process is automatically triggered each time a new document is added, and the index is updated. Specifically, when a new document is added, the server automatically analyzes the document and rebuilds the appropriate index.

[0113] Input: Document information retrieved from the content management system

[0114] Output: Data indexed by ElasticSearch

[0115] Step 2: Enter search keywords

[0116] A user enters keywords into a device to search for specific information. For example, they enter "project management" into the search bar of a web browser. The entered search keywords are sent to the server in JSON format via an AJAX request. Specifically, the device receives the user input, generates a request including the keywords, and sends it to the server.

[0117] Input: The search keyword entered by the user

[0118] Output: Search keyword request in JSON format

[0119] Step 3: Finding relevant documentation

[0120] The server searches the index based on the received search keywords. It generates and executes an ElasticSearch search query to extract the most relevant documents. The extracted document metadata (title, path, etc.) is packaged in JSON format and sent to the device as a response. Specifically, the server generates a search query and queries the indexed data to select relevant documents.

[0121] Input: Search keyword request in JSON format

[0122] Output: Associated document metadata in JSON format

[0123] Step 4: Viewing search results

[0124] The terminal displays the search results received from the server to the user. JavaScript is used to dynamically generate HTML elements and create a search result list. For example, a list of titles such as "Project Management Basics" and "Efficient Project Management Methods" is displayed, with each item in the list providing a link to a detailed page. Specifically, the terminal parses the received JSON data, generates the necessary HTML structure, and inserts it into the page.

[0125] Input: Associated document metadata in JSON format

[0126] Output: A list of search results that the user can see in their browser

[0127] Step 5: Record and learn from your search history

[0128] The server records the user's search history. Specifically, it saves search keywords and search execution timestamps in a NoSQL database such as MongoDB. This data is analyzed by a machine learning model (TensorFlow or Scikit-learn) to learn the user's search patterns. This makes it possible to provide more personalized results the next time the user searches. Specifically, the server retrieves the recorded search history and uses a machine learning algorithm to generate a training dataset and update the model.

[0129] Input: Search keyword and timestamp

[0130] Output: Machine learning model updates

[0131] Step 6: Use the generative AI model

[0132] The server uses a generative AI model to understand the context of the entered search keywords. For example, it uses OpenAI's GPT-3 model to generate context-analyzing prompts. The generative AI model analyzes the meaning of the keywords and generates advanced search results. Specifically, the server sends prompts to the GPT-3 API, analyzes the results, and filters and re-evaluates relevant documents.

[0133] Input: Search keywords and prompt text

[0134] Output: Contextualized, advanced search results

[0135] The above is the specific processing flow of this system, explaining what kind of data processing and data calculation is performed at each step and what kind of output is obtained as a result.

[0136] (Application example 1)

[0137] 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."

[0138] Current information search systems in content distribution services have difficulty quickly and accurately searching for the necessary information from a large amount of information and providing personalized search results based on the user's search history. Another problem is the lack of advanced search functions that utilize generative AI models and functions that learn the user's search trends to optimize search results for future searches.

[0139] 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.

[0140] In this invention, the server includes a means for indexing necessary information from a large amount of information, a means for receiving search keywords from a user, and a means for extracting related documents from the indexed data based on the search keywords. This adds a function for recommending personalized content using the user's search history and a machine learning model, enabling fast and accurate information retrieval. Furthermore, the generative AI model can be used to understand context and generate more appropriate search results, improving user convenience.

[0141] text

[0142] "A large amount of information" refers to a large number of documents and data, indicating a huge amount of information.

[0143] "Indexing" is the process of pre-organizing information to make searching more efficient and providing quick access to key data.

[0144] "Search keywords" refer to specific words or phrases that users enter to search for information.

[0145] "Related documents" refer to documents with related content that are extracted based on the search keywords.

[0146] "User search history" refers to records of past search queries and their results.

[0147] "Search trends" refer to search patterns and preferences derived from a user's past search history.

[0148] A "machine learning model" refers to an algorithm that learns from large amounts of data and makes inferences and predictions.

[0149] A "generative AI model" is an artificial intelligence model that understands the context of the search keywords entered and generates optimal results based on that.

[0150] A "content distribution service" refers to a service that provides users with various content such as videos, articles, and music via the Internet.

[0151] "Personalized content" refers to content that is individually tailored to a user based on their preferences and past behavior.

[0152] "Recommendation methods" refer to methods of selecting and presenting appropriate content based on a user's search history and trends.

[0153]

[0154] This invention is a system for quickly and accurately searching for necessary information from a large amount of information and providing personalized search results based on the user's search history. In particular, this system enhances the personalized content recommendation function in content distribution services by utilizing the user's search history and machine learning models.

[0155] Hardware and Software Configuration

[0156] 1. Server: The server works with multiple databases to index and store large amounts of documents. The indexed data is managed using a search engine such as Elasticsearch.

[0157] 2. User device: User devices such as smartphones and head-mounted displays (HMDs) receive search keywords from users and send them to the server. They also display search results and recommended content to users.

[0158] 3. Generative AI Model: The server uses a generative AI model (e.g., Hugging Face Transformers) to understand the context of the search keywords and generate optimal search results based on that. This generative AI model is also used to update the search history and personalized models.

[0159] System Operation

[0160] 1. Data Indexing:

[0161] The server retrieves large amounts of document information and indexes it using Elasticsearch, which is automatically updated whenever new documents are added.

[0162] 2. Receiving search keywords:

[0163] Users input search keywords via their smartphone or HMD, which are then sent to the server.

[0164] 3. Search for related documents:

[0165] The server quickly extracts relevant documents from the indexed data based on search keywords, and the extracted documents are personalized based on the user's search history.

[0166] 4. Displaying search results:

[0167] The user terminal displays the received search results to the user, including document titles and links that the user can refer to to access more detailed information.

[0168] 5. Learn your search history:

[0169] The server records the user's search history and updates the machine learning model based on this information, allowing the user's search trends to be reflected in subsequent searches, resulting in more relevant search results.

[0170] 6. Personalized Recommendations:

[0171] The server uses a generative AI model to recommend optimal content based on the user's search history and behavioral patterns, allowing users to quickly access new content based on their past search history and behavior.

[0172] Examples and prompts

[0173] A specific example is when a user searches for "the latest trends in project management" on a smartphone. In this case, the user types "the latest trends in project management."

[0174] Examples of prompts:

[0175] "Latest Trends in Project Management"

[0176] This prompt is sent to the server, which then extracts relevant documents from the indexed data and sends them to the user's device. The device displays them to the user, allowing them to quickly access the information they need. This search history is also recorded and reflected in future searches.

[0177] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0178] Step 1:

[0179] The server retrieves large amounts of document information. Specifically, it retrieves document data from a content management system and indexes it using a search engine such as Elasticsearch. It receives document information as input and outputs it as indexed data. This process is performed to reduce search time for search queries.

[0180] Step 2:

[0181] The user inputs search keywords through a smartphone or head-mounted display (HMD). The input keywords are sent from the user device to the server. In this step, the input information about what the user is looking for is collected and becomes input data for the next processing step.

[0182] Step 3:

[0183] The server searches the indexed data based on the received search keywords, using Elasticsearch to quickly extract documents related to the search keywords. The input is the search keywords submitted by the user and the indexed data, and the output is a list of relevant documents.

[0184] Step 4:

[0185] The server uses a generative AI model to understand the context of the search keywords and generate optimal search results based on that. Specifically, it uses a generative AI model, such as Hugging Face's Transformers, to evaluate related documents and select the best results. In this step, the search keywords and related documents are used as input, and personalized search results are generated as output.

[0186] Step 5:

[0187] The server sends the generated search results to the user terminal. The user terminal displays the received search results to the user. The displayed information includes document titles and links, allowing the user to quickly access the information they need. The input of this step is the generated search results, and the output is the search results displayed to the user.

[0188] Step 6:

[0189] The server records the user's search history, specifically the search queries and documents selected by the user, and stores them in a database. The input is the user's history of actions, and the output is a stored search history. This data is used to optimize future searches.

[0190] Step 7:

[0191] The server updates the machine learning model based on the search history. Specifically, it uses the saved search history to train the model to reflect the user's preferences in the next search. The input for this step is the user's search history, and the output is an updated machine learning model.

[0192] Step 8:

[0193] The server then uses the updated machine learning model to recommend personalized content based on the user's search history and behavioral patterns. In this step, the user's search history and the machine learning model are used as inputs, and a personalized content list is generated as output.

[0194] As described above, the system goes through a series of processing steps to provide users with a fast and personalized information retrieval experience.

[0195] 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.

[0196] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends and emotions. This system works in cooperation with a server, a terminal, and a user.

[0197] Data Indexing

[0198] The server first retrieves all document information from the content management system and indexes them. The indexed data is used to speed up searches and is stored in an index data store. Indexing occurs automatically whenever a new document is added.

[0199] Enter search keywords and sentiment data

[0200] Users input search keywords into their devices to search for specific information. The devices also have an emotion engine that recognizes the user's emotions, and collects the user's emotion data.

[0201] Submitting a search request

[0202] The terminal receives search keywords and emotion data input by the user and transmits this information to the server.

[0203] Retrieving relevant documents and using sentiment data

[0204] The server searches the index based on the received search keywords to extract the most relevant documents, taking into account the user's emotional data provided by the emotion engine and optimizing the search results in a format appropriate to the user's emotions.

[0205] Displaying search results

[0206] The terminal displays the search results received from the server to the user. The displayed search results include titles and paths, which the user can refer to to access the desired information.

[0207] Recording and learning search history and emotion data

[0208] The server records the user's search history and emotion data, associates them with the user profile, and stores them. Furthermore, by updating the machine learning model, the server can provide more appropriate search results based on the user's search habits and emotions the next time the user searches.

[0209] Using generative AI models

[0210] The server uses a generative AI model to understand the context of the user's search keywords and generate optimal search results, which also take into account sentiment data.

[0211] Specific examples

[0212] For example, consider a case where a user searches for information about "project management." The user types "project management" into the device. At the same time, the emotion engine detects that the user is feeling stressed. The device sends the search keywords and emotion data to the server. The server searches the index based on the keyword "project management" and extracts relevant documents.

[0213] As a search result, the server extracts documents with titles such as "Project Management Fundamentals" and "Efficient Project Management Methods" and sends them to the terminal. The terminal displays them to the user, allowing the user to quickly access the information they need.

[0214] Additionally, the server records the user's search history and emotional data, and learns to prioritize information related to "project management" and their emotional state in future searches. If the user is feeling stressed, the server may also display information on relaxation techniques and articles that can help reduce stress.

[0215] This system combines a user's search history with emotional data to provide more relevant information that is in line with the user's emotions quickly and accurately, improving the information search experience and ultimately increasing work efficiency and productivity.

[0216] The processing flow will be explained below.

[0217] Step 1:

[0218] The server retrieves all document information from the content management system and indexes it. The indexed data is stored in an index data store for fast searching. This index is automatically updated whenever new documents are added.

[0219] Step 2:

[0220] A user inputs search keywords into a device to search for specific information. The device, which is equipped with an emotion engine, simultaneously recognizes the user's emotional state using technologies such as voice analysis and facial expression recognition.

[0221] Step 3:

[0222] The terminal receives search keywords and emotion data from the user and transmits this information to the server in a transmission format that includes the search keywords and emotion data.

[0223] Step 4:

[0224] The server searches for relevant documents from the index data store based on the received search keywords. The search query is analyzed by a parser to extract the most relevant documents. At the same time, the search results are optimized based on sentiment data.

[0225] Step 5:

[0226] The server then aggregates the extracted relevant document information into search results that take into account the user's emotional state and are formatted to best fit the user's current emotions.

[0227] Step 6:

[0228] The server sends the formatted search results to the terminal, which include the document title and path.

[0229] Step 7:

[0230] The terminal displays the search results received from the server to the user, who can then refer to the displayed results and access the information they need.

[0231] Step 8:

[0232] The server records the user's search queries and sentiment data, which are then associated with the user's profile to aid in future searches.

[0233] Step 9:

[0234] The server uses the recorded search history and sentiment data to update the machine learning model, which is then used to provide more accurate search results the next time the user searches.

[0235] Step 10:

[0236] The server uses a generative AI model to understand the context of the user's search keywords and generate optimal search results, which also take into account sentiment data.

[0237] Step 11:

[0238] Users will get faster, more relevant search results on subsequent searches, improving their information-seeking experience and overall business efficiency.

[0239] Specific examples

[0240] For example, if a user searches for information about "project management," the user enters "project management" into the device. The emotion engine detects stress from the user's tone of voice. The device then sends the search keyword "project management" and emotion data (stress state) to the server.

[0241] The server searches the index data store based on the received search keyword "project management" and extracts relevant documents. In addition, taking into account the emotional data indicating that the user is feeling stressed, documents related to stress management and relaxation methods are also included in the search results.

[0242] As search results, the server extracts documents with titles such as "Project Management Basics" and "Efficient Project Management Methods," as well as documents such as "Methods for managing stress in project management," and sends these to the terminal.

[0243] The device displays the results to the user, allowing them to quickly and appropriately access the information they need. The server records this search history and sentiment data and updates the machine learning model to help with future searches.

[0244] In this way, Knowledge Comet combines users' search history and sentiment data to provide more relevant and contextually appropriate information quickly and accurately, resulting in an improved information search experience and increased work efficiency and productivity.

[0245] Example 2

[0246] 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."

[0247] In today's world, where a vast amount of information is digitized, it remains difficult for users to quickly and accurately search for the information they need. Conventional search systems only provide simple keyword-based search functions and are unable to fully consider users' emotions and search trends. As a result, users' needs cannot be fully met, resulting in a limited search experience. Furthermore, search results that are not tailored to users' emotions can cause frustration and dissatisfaction. The present invention aims to address these issues and improve the information search experience by providing personalized search results that take into account users' emotions and search history.

[0248] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0249] In this invention, the server includes: a means for indexing necessary information from a large amount of information; a means for receiving search keywords from a user; a means for analyzing the user's emotions using collected emotion data; a means for extracting related documents from the indexed data based on the search keywords and emotion data; a means for optimizing search results in a format appropriate for the user's emotions and displaying them to the user; a means for recording the user's search history and emotion data and learning search trends; and a means for generating optimal search results using a generative AI model. This enables the provision of search results that take into account not only the user's search keywords but also the user's emotional state, resulting in a more satisfying information search experience for the user. Furthermore, continuous learning using the machine learning model can improve the accuracy of search results from the next search onwards.

[0250] "Indexing" is the process of organizing data and information according to specific rules and attaching tags and keys to make searching more efficient.

[0251] "Search keywords" are words or phrases that a user enters to search for specific information.

[0252] "Emotion data" refers to information that expresses the user's emotional state using numerical values ​​or categories, and includes, for example, joy, sadness, surprise, and the like.

[0253] "Related documents" are documents or information that are determined to be most relevant based on the search keywords and emotion data entered by the user.

[0254] "Optimization" means adjusting something to the most effective or efficient state for specific conditions or goals.

[0255] "Search history" refers to a record of a user's past search queries, links they have clicked, and so on.

[0256] A "generative AI model" is artificial intelligence that uses advanced algorithms, such as deep learning, to generate text and other forms of data.

[0257] A "machine learning model" is an algorithm or statistical model that is trained on data to automatically perform a specific task.

[0258] A "user profile" is a data set that aggregates a user's basic information, search history, emotional data, etc.

[0259] "Natural language processing" is a technology that allows machines to understand, analyze, and generate human language.

[0260] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends and emotions. This system works in cooperation with a server, a terminal, and a user.

[0261] Data Indexing

[0262] The server first retrieves all document information from the content management system and indexes it. Indexing is done using the Apache Lucene search engine, and the results are stored in MongoDB. The indexed data is used to speed up searches and is automatically updated whenever new documents are added.

[0263] Enter search keywords and sentiment data

[0264] A user inputs search keywords into a device to search for specific information. The device is equipped with a camera that analyzes the user's facial expressions and a microphone that analyzes the tone of the user's voice, and collects the user's emotional data using an emotion recognition API (for example, Microsoft Azure's emotion recognition API).

[0265] Submitting a search request

[0266] The device receives search keywords and emotion data entered by the user and sends this information to the server. The request is constructed in JSON format using JavaScript and sent to the server as an HTTP POST request.

[0267] Retrieving relevant documents and using sentiment data

[0268] The server analyzes the received search keywords and sentiment data, searches the index data store, extracts the most relevant documents, and feeds the results into a generative AI model (e.g., OpenAI's GPT-3) to generate search results that fit the user's sentiment.

[0269] Displaying search results

[0270] The terminal renders the search results received from the server in HTML format and displays them in the browser. The search results include document titles and access paths, which users can refer to to access the desired information. The terminal uses React.js.

[0271] Recording and learning search history and emotion data

[0272] The server records the user's search history and sentiment data in a database (e.g., Apache Hadoop) and continuously updates a machine learning model (e.g., Scikit-learn). This data is associated with the user's profile and used to provide better search results the next time the user searches.

[0273] Examples and prompts

[0274] As a concrete example, consider a case where a user searches for information on "project management." In this case, the user enters "project management" into the device, and the emotion engine detects that the user is feeling stressed. The device sends the search keywords and emotion data to the server, which then extracts related documents based on the keyword "project management." In this case, a generative AI model is used to generate search results, and documents with titles such as "Project Management Basics" and "Efficient Project Management Methods" are sent to the device. The device displays these results to the user, allowing them to quickly access the information they need.

[0275] Example prompt sentence:

[0276] "In a search system that is integrated with an emotion engine, if a user searches for the keyword 'project management' and detects that they are feeling stressed, what search results should be displayed? Use a generative AI model to suggest the most appropriate search results."

[0277] This system combines a user's search history with emotional data to provide more relevant information that is in line with the user's emotions quickly and accurately, improving the information search experience. Furthermore, continuous learning using machine learning models improves the accuracy of search results from the next time onwards.

[0278] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0279] Step 1:

[0280] Data Indexing

[0281] The server retrieves all document information from the content management system. The input is the document information from the content management system. The server parses and indexes the document information using Apache Lucene. The indexed data is stored in MongoDB, providing an index data store for efficient data searches. When new documents are added, the index is automatically updated.

[0282] Step 2:

[0283] Enter search keywords and sentiment data

[0284] The user enters a search keyword (e.g., "project management") into the search bar and sends it to the device. The input at this time is the search keyword entered by the user. The device uses the built-in camera and microphone to analyze the user's facial expressions and tone of voice to collect emotional data. The emotional data is analyzed using Microsoft Azure's emotion recognition API. This results in the search keyword and the user's emotional data.

[0285] Step 3:

[0286] Submitting a search request

[0287] The device combines the acquired search keywords and emotion data to construct a search request. The input at this time is the search keywords and emotion data. The device uses JavaScript to form the request in JSON format and sends it to the server as an HTTP POST request. This results in a search request to the server.

[0288] Step 4:

[0289] Retrieving relevant documents and using sentiment data

[0290] The server analyzes the received search request and extracts search keywords and sentiment data. The input is the search request, and the output is the search keywords and sentiment data. The server searches the index data store and extracts the most relevant documents based on the search keywords. A generative AI model (e.g., OpenAI's GPT-3) is used for this data extraction. The generative AI model takes sentiment data into account to generate optimal search results, resulting in the most relevant search results.

[0291] Step 5:

[0292] Displaying search results

[0293] The server sends the generated search results to the terminal. The input at this time is the generated search results, and the output is the search result response from the server. The terminal renders the received search results in HTML format and displays them to the user. React.js is used for display. This allows the user to quickly access the desired information by referring to the titles and access paths of related documents.

[0294] Step 6:

[0295] Recording and learning search history and emotion data

[0296] The server records the user's search history and sentiment data in a database (e.g., Apache Hadoop). The input is the user's search history and sentiment data. The output is the search history and sentiment data after it has been saved in the database. This data is used to improve the accuracy of search results from the next time onwards using a machine learning model (e.g., Scikit-learn). This updates the user profile, and more relevant results will be provided for the next search.

[0297] Step 7:

[0298] Using generative AI models

[0299] The server uses a generative AI model to understand the context of the user's search keywords and emotional data. The inputs are the search keywords and emotional data. The output is search results optimized based on the context. The generative AI model (e.g., OpenAI's GPT-3) generates search results that also take emotional data into account. This provides search results that are appropriate for the user's emotions.

[0300] (Application example 2)

[0301] 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."

[0302] Conventional information search systems have difficulty quickly and appropriately providing the necessary information from a large amount of information. Furthermore, they lacked personalization functionality that took into account the user's search trends and emotional state, making it impossible to provide the most useful information for the user. As a result, information search efficiency declined, and user satisfaction was compromised.

[0303] 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 indexing necessary information from a large amount of information, means for receiving search keywords from the user, means for extracting related documents from the indexed data based on the search keywords, means for displaying search results to the user, means for recording the user's search history and emotional data and learning the user's search tendencies and emotional state, and means for personalizing search results based on the user's emotional data. This makes it possible to quickly and accurately provide information related to the user's search keywords and emotional data.

[0304] "A means of indexing necessary information from a large amount of information" is the process by which a server retrieves all document information from a content management system and stores it in an index data store to speed up searches.

[0305] The "means for receiving search keywords from a user" refers to a process in which a user inputs search keywords to search for specific information through a terminal.

[0306] "Means for extracting relevant documents from indexed data based on search keywords" refers to the process by which the server searches the index based on the search keywords and extracts the most relevant documents.

[0307] The "means for displaying search results to the user" refers to the process by which the terminal displays the search results received from the server on the screen, allowing the user to access the desired information.

[0308] "Means for recording users' search history and emotional data and learning their search tendencies and emotional states" refers to the process by which the server records users' past search history and emotional data, and updates the algorithm based on this data to understand users' search tendencies and emotions.

[0309] "Means for personalizing search results based on user emotional data" refers to a process in which the server utilizes the user's emotional data obtained from the emotion engine to optimize search results to adapt to the user's emotional state.

[0310] "Means for updating the machine learning model based on search query input and sentiment data to reflect the user's tendencies and emotional state for future searches" refers to the process by which the server uses the user's search query and sentiment data to train the machine learning model to provide results that adapt to the user's individual tendencies and emotional state for future searches.

[0311] "Means of using a generative AI model to quickly and accurately provide information related to a user's search keywords and sentiment data" refers to the process by which a generative AI model understands a user's search keywords and sentiment data and generates optimal search results.

[0312] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends and emotions. This system works in cooperation with a server, a terminal, and an emotion engine.

[0313] The server first retrieves all document information from the content management system and indexes them. The indexed data is used to speed up searches and is stored in an index data store. Indexing occurs automatically whenever a new document is added.

[0314] To search for specific information, a user inputs search keywords into the device. The device also has an emotion engine that recognizes the user's emotions, and collects the user's emotion data.

[0315] The terminal receives search keywords and emotion data input by the user and transmits this information to the server.

[0316] The server searches the index based on the received search keywords to extract the most relevant documents, taking into account the user's emotional data and optimizing the search results in a format that is appropriate for the user's emotions.

[0317] The terminal displays the search results received from the server to the user. The displayed search results include titles and paths, which the user can refer to to access the desired information.

[0318] The server records the user's search history and emotion data, associates them with the user profile, and stores them. Furthermore, by updating the machine learning model, the server can provide more relevant search results based on the user's search habits and emotions the next time the user searches.

[0319] The server uses a generative AI model to generate optimal search results based on the user's search keywords and emotional data. Specifically, the generative AI model can generate search results taking into account the user's emotional data.

[0320] For example, consider a case where a user searches for "soothing music." The user enters "soothing music" into the device, and the emotion engine detects "stress." The device sends the search keywords and emotion data to the server. The server extracts documents related to "soothing music" from the indexed data and provides a list of music that helps reduce stress. In this case, the generative AI model uses prompts such as the following to generate optimal search results:

[0321] Example prompt sentence:

[0322] "A user searches for 'soothing music'. The user's emotional state is 'stressed'. Generate a list of music that will help reduce stress."

[0323] This makes it possible to provide information quickly and accurately according to the user's emotional state and search tendencies, thereby realizing a system that contributes to improving user satisfaction.

[0324] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0325] Step 1:

[0326] The server retrieves all document information from the content management system and indexes it. This indexed data is used for faster searches and is stored in an index data store. The input is the document data from the content management system and the output is the indexed data.

[0327] Step 2:

[0328] To search for specific information, a user inputs search keywords into the device. The device also incorporates an emotion engine that recognizes the user's emotions, and simultaneously collects the user's emotional data. The input is the user's search keywords and emotional data, and the output is the integrated information of these data.

[0329] Step 3:

[0330] The terminal receives search keywords and emotion data entered by the user and transmits this information to the server. The input is the search keywords and emotion data, and the output is the data transmission result to the server.

[0331] Step 4:

[0332] The server searches the index data store based on the received search keywords to extract the most relevant documents, taking into account the user's emotional data and optimizing the search results in a format that is appropriate for the user's emotions. The input is the search keywords and emotional data, and the output is personalized search results.

[0333] Step 5:

[0334] The terminal displays the search results received from the server to the user. The displayed search results include a title and a path, which the user can refer to to access the desired information. The input is the search results from the server, and the output is the search results displayed to the user.

[0335] Step 6:

[0336] The server records the user's search history and emotion data, associates them with the user profile, and saves them. Furthermore, by updating the machine learning model, the server can provide more appropriate search results based on the user's search trends and emotions the next time they search. The input is the search history and emotion data, and the output is the updated machine learning model.

[0337] Step 7:

[0338] The server uses a generative AI model to generate optimal search results based on the user's search keywords and emotional data. Specifically, the generative AI model can generate search results taking into account the user's emotional data. The input is the search keywords, emotional data, and the generative AI model's prompt, and the output is personalized search results.

[0339] For example, if a user searches for "soothing music," an example prompt might look like this:

[0340] "A user searches for 'soothing music'. The user's emotional state is 'stressed'. Generate a list of music that will help reduce stress."

[0341] 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.

[0342] 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.

[0343] 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.

[0344] [Second embodiment]

[0345] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0346] 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.

[0347] 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).

[0348] 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.

[0349] 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.

[0350] 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).

[0351] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0352] 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.

[0353] 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.

[0354] 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.

[0355] 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.

[0356] 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."

[0357] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends. This system works in cooperation with a server, a terminal, and a user.

[0358] Data Indexing

[0359] The server first retrieves all document information from the content management system and indexes them. The indexed data is used to speed up searches. Indexing occurs automatically whenever a new document is added.

[0360] Enter search keywords

[0361] A user inputs search keywords into a terminal to search for specific information, and the search keywords are transmitted to a server via the terminal.

[0362] Find related documents

[0363] The server searches the index based on the received search keywords to extract the most relevant documents, using rapid references to the indexed data to generate optimal search results.

[0364] Displaying search results

[0365] The terminal displays the search results received from the server to the user. The displayed search results include titles and paths, which the user can refer to to access the desired information.

[0366] Record and learn from your search history

[0367] The server records the user's search history and updates the machine learning model based on this information, allowing the user's search trends to be reflected in the next search, resulting in more relevant search results.

[0368] Using generative AI models

[0369] Additionally, the server uses a generative AI model to understand the context of the search keywords entered and generate optimal search results, which goes beyond simple keyword matching to achieve more advanced searches.

[0370] Specific examples

[0371] For example, consider a case where a user searches for information about "project management." The user types "project management" into a terminal. The terminal sends this input to a server. The server searches the index based on the keyword "project management" and extracts relevant documents.

[0372] As a search result, the server extracts documents with titles such as "Project Management Fundamentals" and "Efficient Project Management Methods" and sends them to the terminal. The terminal displays them to the user, allowing the user to quickly access the information they need.

[0373] The server also records the user's search history and learns to prioritize displaying information related to "project management" in subsequent searches.

[0374] By combining users' search history with generative AI models, this system can provide more accurate and personalized search results, significantly reducing the time required to search for information.

[0375] In this way, Knowledge Comet provides users with a fast and accurate information search experience, contributing to improved overall business efficiency and productivity.

[0376] The processing flow will be explained below.

[0377] Step 1:

[0378] The server retrieves all document information from the content management system, indexes these documents, transforms the indexed data into a structure that facilitates fast searches, and stores it in an index data store.

[0379] Step 2:

[0380] A user inputs search keywords into a terminal to search for specific information. The search keywords input by a user are specific words or phrases.

[0381] Step 3:

[0382] The terminal receives search keywords entered by the user and transmits this information to the server, where the search keywords are processed.

[0383] Step 4:

[0384] The server takes the received search keywords and searches for relevant documents from the index data store. The search query is analyzed by a parser to extract the most relevant documents.

[0385] Step 5:

[0386] The server compiles the extracted related document information (e.g., title, path) as search results, and formats the search results in a user-friendly format.

[0387] Step 6:

[0388] The server sends the formatted search results to the terminal, which include the document title and path.

[0389] Step 7:

[0390] The terminal displays the search results received from the server to the user, who can then refer to the displayed results and access the information they need.

[0391] Step 8:

[0392] The server records the user's search queries and stores them as the user's search history, which is associated with the user's profile.

[0393] Step 9:

[0394] The server uses the saved search history to update the machine learning model, which is then used to provide better search results the next time the user searches.

[0395] Step 10:

[0396] The server uses a generative AI model to understand the context of the user's search keywords, generating advanced search results that go beyond simple keyword matching.

[0397] Step 11:

[0398] Users will be able to get more relevant search results faster on subsequent searches, improving their information-seeking experience.

[0399] Example 1

[0400] 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."

[0401] Conventional information retrieval systems have difficulty quickly and accurately searching for the necessary information from a large amount of information, and it is difficult to provide personalized search results that appropriately reflect the user's search trends. Furthermore, because they rely on simple keyword matching, they are unable to realize advanced searches that understand the context. Furthermore, when a user conducts multiple searches, the system lacks the ability to fully learn the user's search history and utilize it for future searches.

[0402] 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.

[0403] In this invention, the server includes: means for indexing necessary information from a large amount of information; means for receiving search keywords from a user; means for extracting related data from the indexed data based on the search keywords; means for displaying search results to the user; means for recording the user's search history and learning search trends; and means for analyzing the context of the entered search keywords using a generative AI model to generate optimal search results. By combining the user's search history and the generative AI model, it is possible to provide more accurate and personalized search results and significantly reduce the time required for information search.

[0404] "Indexing" is the process of organizing and classifying large amounts of information based on specific criteria so that it can be easily searched.

[0405] A "search keyword" is a word or phrase that a user enters when searching for specific information.

[0406] "Related data" refers to data items related to the targeted information that are extracted based on the user's search keywords.

[0407] "Search results" are a list of related data returned in response to a user's search keywords and displayed to the user.

[0408] "Search history" is a record of searches that a user has performed up to now, and includes information such as date and time and keywords.

[0409] "Search trends" are patterns and characteristics extracted from a user's past search behavior.

[0410] A "generative AI model" is a mathematical model that uses artificial intelligence to analyze the context and meaning of input data and generate appropriate output.

[0411] "Context" is other information or data that surrounds a keyword or statement and influences its meaning and interpretation.

[0412] "Personalization" means providing services and content that are specifically tailored and optimized based on individual user characteristics and behavior.

[0413] "Information retrieval" is the process of extracting specific information from a particular database or index.

[0414] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends. This system works in cooperation with a server, a terminal, and a user.

[0415] Data Indexing

[0416] The server first retrieves all document information from the content management system and indexes it. This process uses search engine software such as ElasticSearch or Apache Lucene. Every time a new document is added, the server automatically updates the index, speeding up searches.

[0417] Enter search keywords

[0418] A user inputs keywords into the device to search for specific information. For example, if a user inputs the keyword "project management," the device sends this input to the server via a REST API. The search keywords are sent to the server in JSON format using an AJAX request.

[0419] Find related documents

[0420] The server searches the index based on the received search keywords, generates and executes an ElasticSearch search query to extract the most relevant documents, and then packages the extracted document metadata (title, path, etc.) in JSON format and sends it to the device as a response.

[0421] Displaying search results

[0422] The terminal displays the search results received from the server to the user. Specifically, JavaScript is used to dynamically generate HTML elements and create a search result list. For example, the search results may display titles such as "Project Management Basics" and "Efficient Project Management Methods," with each title serving as a link to the corresponding document's detailed page.

[0423] Record and learn from your search history

[0424] The server records the user's search history. Specifically, it stores the search keywords and search execution timestamps in MongoDB, a NoSQL database. This recorded data is used to update machine learning models (e.g., TensorFlow and Scikit-learn) and learn the user's search patterns. This allows the server to provide more personalized search results the next time the user searches.

[0425] Using generative AI models

[0426] Furthermore, the server uses a generative AI model to understand the context of the search keywords entered. For example, by using OpenAI's GPT-3 model, it achieves a more advanced search than simple keyword matching. The generative AI model understands the context and generates advanced search results. Specific prompts could be something like:

[0427] "What are some basic project management information?"

[0428] "I'd like to get an overview of efficient project management methods."

[0429] Specific examples

[0430] For example, a user may search for information related to "project management." The user types "project management" into their device, which then sends this input to the server via an AJAX request. The server uses ElasticSearch to extract relevant documents from the index and returns the document information to the device. The device displays the search results to the user, who can click links to access more information. The server then records this search history and learns to prioritize displaying information related to "project management" in future searches.

[0431] By combining users' search history with generative AI models, the system can provide more accurate and personalized search results, significantly reducing the time it takes to find information.

[0432] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0433] Step 1: Index the data

[0434] The server first retrieves all document information from the content management system. This is done using SQL queries or API requests. To index the retrieved document information, ElasticSearch is used to organize and classify the data and create an index that enables fast searches. This process is automatically triggered each time a new document is added, and the index is updated. Specifically, when a new document is added, the server automatically analyzes the document and rebuilds the appropriate index.

[0435] Input: Document information retrieved from the content management system

[0436] Output: Data indexed by ElasticSearch

[0437] Step 2: Enter search keywords

[0438] A user enters keywords into a device to search for specific information. For example, they enter "project management" into the search bar of a web browser. The entered search keywords are sent to the server in JSON format via an AJAX request. Specifically, the device receives the user input, generates a request including the keywords, and sends it to the server.

[0439] Input: The search keyword entered by the user

[0440] Output: Search keyword request in JSON format

[0441] Step 3: Finding relevant documentation

[0442] The server searches the index based on the received search keywords. It generates and executes an ElasticSearch search query to extract the most relevant documents. The extracted document metadata (title, path, etc.) is packaged in JSON format and sent to the device as a response. Specifically, the server generates a search query and queries the indexed data to select relevant documents.

[0443] Input: Search keyword request in JSON format

[0444] Output: Associated document metadata in JSON format

[0445] Step 4: Viewing search results

[0446] The terminal displays the search results received from the server to the user. JavaScript is used to dynamically generate HTML elements and create a search result list. For example, a list of titles such as "Project Management Basics" and "Efficient Project Management Methods" is displayed, with each item in the list providing a link to a detailed page. Specifically, the terminal parses the received JSON data, generates the necessary HTML structure, and inserts it into the page.

[0447] Input: Associated document metadata in JSON format

[0448] Output: A list of search results that the user can see in their browser

[0449] Step 5: Record and learn from your search history

[0450] The server records the user's search history. Specifically, it saves search keywords and search execution timestamps in a NoSQL database such as MongoDB. This data is analyzed by a machine learning model (TensorFlow or Scikit-learn) to learn the user's search patterns. This makes it possible to provide more personalized results the next time the user searches. Specifically, the server retrieves the recorded search history and uses a machine learning algorithm to generate a training dataset and update the model.

[0451] Input: Search keyword and timestamp

[0452] Output: Machine learning model updates

[0453] Step 6: Use the generative AI model

[0454] The server uses a generative AI model to understand the context of the entered search keywords. For example, it uses OpenAI's GPT-3 model to generate context-analyzing prompts. The generative AI model analyzes the meaning of the keywords and generates advanced search results. Specifically, the server sends prompts to the GPT-3 API, analyzes the results, and filters and re-evaluates relevant documents.

[0455] Input: Search keywords and prompt text

[0456] Output: Contextualized, advanced search results

[0457] The above is the specific processing flow of this system, explaining what kind of data processing and data calculation is performed at each step and what kind of output is obtained as a result.

[0458] (Application example 1)

[0459] 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."

[0460] Current information search systems in content distribution services have difficulty quickly and accurately searching for the necessary information from a large amount of information and providing personalized search results based on the user's search history. Another problem is the lack of advanced search functions that utilize generative AI models and functions that learn the user's search trends to optimize search results for future searches.

[0461] 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.

[0462] In this invention, the server includes a means for indexing necessary information from a large amount of information, a means for receiving search keywords from a user, and a means for extracting related documents from the indexed data based on the search keywords. This adds a function for recommending personalized content using the user's search history and a machine learning model, enabling fast and accurate information retrieval. Furthermore, the generative AI model can be used to understand context and generate more appropriate search results, improving user convenience.

[0463] text

[0464] "A large amount of information" refers to a large number of documents and data, indicating a huge amount of information.

[0465] "Indexing" is the process of pre-organizing information to make searching more efficient and providing quick access to key data.

[0466] "Search keywords" refer to specific words or phrases that users enter to search for information.

[0467] "Related documents" refer to documents with related content that are extracted based on the search keywords.

[0468] "User search history" refers to records of past search queries and their results.

[0469] "Search trends" refer to search patterns and preferences derived from a user's past search history.

[0470] A "machine learning model" refers to an algorithm that learns from large amounts of data and makes inferences and predictions.

[0471] A "generative AI model" is an artificial intelligence model that understands the context of the search keywords entered and generates optimal results based on that.

[0472] A "content distribution service" refers to a service that provides users with various content such as videos, articles, and music via the Internet.

[0473] "Personalized content" refers to content that is individually tailored to a user based on their preferences and past behavior.

[0474] "Recommendation methods" refer to methods of selecting and presenting appropriate content based on a user's search history and trends.

[0475]

[0476] This invention is a system for quickly and accurately searching for necessary information from a large amount of information and providing personalized search results based on the user's search history. In particular, this system enhances the personalized content recommendation function in content distribution services by utilizing the user's search history and machine learning models.

[0477] Hardware and Software Configuration

[0478] 1. Server: The server works with multiple databases to index and store large amounts of documents. The indexed data is managed using a search engine such as Elasticsearch.

[0479] 2. User device: User devices such as smartphones and head-mounted displays (HMDs) receive search keywords from users and send them to the server. They also display search results and recommended content to users.

[0480] 3. Generative AI Model: The server uses a generative AI model (e.g., Hugging Face Transformers) to understand the context of the search keywords and generate optimal search results based on that. This generative AI model is also used to update the search history and personalized models.

[0481] System Operation

[0482] 1. Data Indexing:

[0483] The server retrieves large amounts of document information and indexes it using Elasticsearch, which is automatically updated whenever new documents are added.

[0484] 2. Receiving search keywords:

[0485] Users input search keywords via their smartphone or HMD, which are then sent to the server.

[0486] 3. Search for related documents:

[0487] The server quickly extracts relevant documents from the indexed data based on search keywords, and the extracted documents are personalized based on the user's search history.

[0488] 4. Displaying search results:

[0489] The user terminal displays the received search results to the user, including document titles and links that the user can refer to to access more detailed information.

[0490] 5. Learn your search history:

[0491] The server records the user's search history and updates the machine learning model based on this information, allowing the user's search trends to be reflected in subsequent searches, resulting in more relevant search results.

[0492] 6. Personalized Recommendations:

[0493] The server uses a generative AI model to recommend optimal content based on the user's search history and behavioral patterns, allowing users to quickly access new content based on their past search history and behavior.

[0494] Examples and prompts

[0495] A specific example is when a user searches for "the latest trends in project management" on a smartphone. In this case, the user types "the latest trends in project management."

[0496] Examples of prompts:

[0497] "Latest Trends in Project Management"

[0498] This prompt is sent to the server, which then extracts relevant documents from the indexed data and sends them to the user's device. The device displays them to the user, allowing them to quickly access the information they need. This search history is also recorded and reflected in future searches.

[0499] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0500] Step 1:

[0501] The server retrieves large amounts of document information. Specifically, it retrieves document data from a content management system and indexes it using a search engine such as Elasticsearch. It receives document information as input and outputs it as indexed data. This process is performed to reduce search time for search queries.

[0502] Step 2:

[0503] The user inputs search keywords through a smartphone or head-mounted display (HMD). The input keywords are sent from the user device to the server. In this step, the input information about what the user is looking for is collected and becomes input data for the next processing step.

[0504] Step 3:

[0505] The server searches the indexed data based on the received search keywords, using Elasticsearch to quickly extract documents related to the search keywords. The input is the search keywords submitted by the user and the indexed data, and the output is a list of relevant documents.

[0506] Step 4:

[0507] The server uses a generative AI model to understand the context of the search keywords and generate optimal search results based on that. Specifically, it uses a generative AI model, such as Hugging Face's Transformers, to evaluate related documents and select the best results. In this step, the search keywords and related documents are used as input, and personalized search results are generated as output.

[0508] Step 5:

[0509] The server sends the generated search results to the user terminal. The user terminal displays the received search results to the user. The displayed information includes document titles and links, allowing the user to quickly access the information they need. The input of this step is the generated search results, and the output is the search results displayed to the user.

[0510] Step 6:

[0511] The server records the user's search history, specifically the search queries and documents selected by the user, and stores them in a database. The input is the user's history of actions, and the output is a stored search history. This data is used to optimize future searches.

[0512] Step 7:

[0513] The server updates the machine learning model based on the search history. Specifically, it uses the saved search history to train the model to reflect the user's preferences in the next search. The input for this step is the user's search history, and the output is an updated machine learning model.

[0514] Step 8:

[0515] The server then uses the updated machine learning model to recommend personalized content based on the user's search history and behavioral patterns. In this step, the user's search history and the machine learning model are used as inputs, and a personalized content list is generated as output.

[0516] As described above, the system goes through a series of processing steps to provide users with a fast and personalized information retrieval experience.

[0517] 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.

[0518] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends and emotions. This system works in cooperation with a server, a terminal, and a user.

[0519] Data Indexing

[0520] The server first retrieves all document information from the content management system and indexes them. The indexed data is used to speed up searches and is stored in an index data store. Indexing occurs automatically whenever a new document is added.

[0521] Enter search keywords and sentiment data

[0522] Users input search keywords into their devices to search for specific information. The devices also have an emotion engine that recognizes the user's emotions, and collects the user's emotion data.

[0523] Submitting a search request

[0524] The terminal receives search keywords and emotion data input by the user and transmits this information to the server.

[0525] Retrieving relevant documents and using sentiment data

[0526] The server searches the index based on the received search keywords to extract the most relevant documents, taking into account the user's emotional data provided by the emotion engine and optimizing the search results in a format appropriate to the user's emotions.

[0527] Displaying search results

[0528] The terminal displays the search results received from the server to the user. The displayed search results include titles and paths, which the user can refer to to access the desired information.

[0529] Recording and learning search history and emotion data

[0530] The server records the user's search history and emotion data, associates them with the user profile, and stores them. Furthermore, by updating the machine learning model, the server can provide more appropriate search results based on the user's search habits and emotions the next time the user searches.

[0531] Using generative AI models

[0532] The server uses a generative AI model to understand the context of the user's search keywords and generate optimal search results, which also take into account sentiment data.

[0533] Specific examples

[0534] For example, consider a case where a user searches for information about "project management." The user types "project management" into the device. At the same time, the emotion engine detects that the user is feeling stressed. The device sends the search keywords and emotion data to the server. The server searches the index based on the keyword "project management" and extracts relevant documents.

[0535] As a search result, the server extracts documents with titles such as "Project Management Fundamentals" and "Efficient Project Management Methods" and sends them to the terminal. The terminal displays them to the user, allowing the user to quickly access the information they need.

[0536] Additionally, the server records the user's search history and emotional data, and learns to prioritize information related to "project management" and their emotional state in future searches. If the user is feeling stressed, the server may also display information on relaxation techniques and articles that can help reduce stress.

[0537] This system combines a user's search history with emotional data to provide more relevant information that is in line with the user's emotions quickly and accurately, improving the information search experience and ultimately increasing work efficiency and productivity.

[0538] The processing flow will be explained below.

[0539] Step 1:

[0540] The server retrieves all document information from the content management system and indexes it. The indexed data is stored in an index data store for fast searching. This index is automatically updated whenever new documents are added.

[0541] Step 2:

[0542] A user inputs search keywords into a device to search for specific information. The device, which is equipped with an emotion engine, simultaneously recognizes the user's emotional state using technologies such as voice analysis and facial expression recognition.

[0543] Step 3:

[0544] The terminal receives search keywords and emotion data from the user and transmits this information to the server in a transmission format that includes the search keywords and emotion data.

[0545] Step 4:

[0546] The server searches for relevant documents from the index data store based on the received search keywords. The search query is analyzed by a parser to extract the most relevant documents. At the same time, the search results are optimized based on sentiment data.

[0547] Step 5:

[0548] The server then aggregates the extracted relevant document information into search results that take into account the user's emotional state and are formatted to best fit the user's current emotions.

[0549] Step 6:

[0550] The server sends the formatted search results to the terminal, which include the document title and path.

[0551] Step 7:

[0552] The terminal displays the search results received from the server to the user, who can then refer to the displayed results and access the information they need.

[0553] Step 8:

[0554] The server records the user's search queries and sentiment data, which are then associated with the user's profile to aid in future searches.

[0555] Step 9:

[0556] The server uses the recorded search history and sentiment data to update the machine learning model, which is then used to provide more accurate search results the next time the user searches.

[0557] Step 10:

[0558] The server uses a generative AI model to understand the context of the user's search keywords and generate optimal search results, which also take into account sentiment data.

[0559] Step 11:

[0560] Users will get faster, more relevant search results on subsequent searches, improving their information-seeking experience and overall business efficiency.

[0561] Specific examples

[0562] For example, if a user searches for information about "project management," the user enters "project management" into the device. The emotion engine detects stress from the user's tone of voice. The device then sends the search keyword "project management" and emotion data (stress state) to the server.

[0563] The server searches the index data store based on the received search keyword "project management" and extracts relevant documents. In addition, taking into account the emotional data indicating that the user is feeling stressed, documents related to stress management and relaxation methods are also included in the search results.

[0564] As search results, the server extracts documents with titles such as "Project Management Basics" and "Efficient Project Management Methods," as well as documents such as "Methods for managing stress in project management," and sends these to the terminal.

[0565] The device displays the results to the user, allowing them to quickly and appropriately access the information they need. The server records this search history and sentiment data and updates the machine learning model to help with future searches.

[0566] In this way, Knowledge Comet combines users' search history and sentiment data to provide more relevant and contextually appropriate information quickly and accurately, resulting in an improved information search experience and increased work efficiency and productivity.

[0567] Example 2

[0568] 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."

[0569] In today's world, where a vast amount of information is digitized, it remains difficult for users to quickly and accurately search for the information they need. Conventional search systems only provide simple keyword-based search functions and are unable to fully consider users' emotions and search trends. As a result, users' needs cannot be fully met, resulting in a limited search experience. Furthermore, search results that are not tailored to users' emotions can cause frustration and dissatisfaction. The present invention aims to address these issues and improve the information search experience by providing personalized search results that take into account users' emotions and search history.

[0570] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0571] In this invention, the server includes: a means for indexing necessary information from a large amount of information; a means for receiving search keywords from a user; a means for analyzing the user's emotions using collected emotion data; a means for extracting related documents from the indexed data based on the search keywords and emotion data; a means for optimizing search results in a format appropriate for the user's emotions and displaying them to the user; a means for recording the user's search history and emotion data and learning search trends; and a means for generating optimal search results using a generative AI model. This enables the provision of search results that take into account not only the user's search keywords but also the user's emotional state, resulting in a more satisfying information search experience for the user. Furthermore, continuous learning using the machine learning model can improve the accuracy of search results from the next search onwards.

[0572] "Indexing" is the process of organizing data and information according to specific rules and attaching tags and keys to make searching more efficient.

[0573] "Search keywords" are words or phrases that a user enters to search for specific information.

[0574] "Emotion data" refers to information that expresses the user's emotional state using numerical values ​​or categories, and includes, for example, joy, sadness, surprise, and the like.

[0575] "Related documents" are documents or information that are determined to be most relevant based on the search keywords and emotion data entered by the user.

[0576] "Optimization" means adjusting something to the most effective or efficient state for specific conditions or goals.

[0577] "Search history" refers to a record of a user's past search queries, links they have clicked, and so on.

[0578] A "generative AI model" is artificial intelligence that uses advanced algorithms, such as deep learning, to generate text and other forms of data.

[0579] A "machine learning model" is an algorithm or statistical model that is trained on data to automatically perform a specific task.

[0580] A "user profile" is a data set that aggregates a user's basic information, search history, emotional data, etc.

[0581] "Natural language processing" is a technology that allows machines to understand, analyze, and generate human language.

[0582] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends and emotions. This system works in cooperation with a server, a terminal, and a user.

[0583] Data Indexing

[0584] The server first retrieves all document information from the content management system and indexes it. Indexing is done using the Apache Lucene search engine, and the results are stored in MongoDB. The indexed data is used to speed up searches and is automatically updated whenever new documents are added.

[0585] Enter search keywords and sentiment data

[0586] A user inputs search keywords into a device to search for specific information. The device is equipped with a camera that analyzes the user's facial expressions and a microphone that analyzes the tone of the user's voice, and collects the user's emotional data using an emotion recognition API (for example, Microsoft Azure's emotion recognition API).

[0587] Submitting a search request

[0588] The device receives search keywords and emotion data entered by the user and sends this information to the server. The request is constructed in JSON format using JavaScript and sent to the server as an HTTP POST request.

[0589] Retrieving relevant documents and using sentiment data

[0590] The server analyzes the received search keywords and sentiment data, searches the index data store, extracts the most relevant documents, and feeds the results into a generative AI model (e.g., OpenAI's GPT-3) to generate search results that fit the user's sentiment.

[0591] Displaying search results

[0592] The terminal renders the search results received from the server in HTML format and displays them in the browser. The search results include document titles and access paths, which users can refer to to access the desired information. The terminal uses React.js.

[0593] Recording and learning search history and emotion data

[0594] The server records the user's search history and sentiment data in a database (e.g., Apache Hadoop) and continuously updates a machine learning model (e.g., Scikit-learn). This data is associated with the user's profile and used to provide better search results the next time the user searches.

[0595] Examples and prompts

[0596] As a concrete example, consider a case where a user searches for information on "project management." In this case, the user enters "project management" into the device, and the emotion engine detects that the user is feeling stressed. The device sends the search keywords and emotion data to the server, which then extracts related documents based on the keyword "project management." In this case, a generative AI model is used to generate search results, and documents with titles such as "Project Management Basics" and "Efficient Project Management Methods" are sent to the device. The device displays these results to the user, allowing them to quickly access the information they need.

[0597] Example prompt sentence:

[0598] "In a search system that is integrated with an emotion engine, if a user searches for the keyword 'project management' and detects that they are feeling stressed, what search results should be displayed? Use a generative AI model to suggest the most appropriate search results."

[0599] This system combines a user's search history with emotional data to provide more relevant information that is in line with the user's emotions quickly and accurately, improving the information search experience. Furthermore, continuous learning using machine learning models improves the accuracy of search results from the next time onwards.

[0600] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0601] Step 1:

[0602] Data Indexing

[0603] The server retrieves all document information from the content management system. The input is the document information from the content management system. The server parses and indexes the document information using Apache Lucene. The indexed data is stored in MongoDB, providing an index data store for efficient data searches. When new documents are added, the index is automatically updated.

[0604] Step 2:

[0605] Enter search keywords and sentiment data

[0606] The user enters a search keyword (e.g., "project management") into the search bar and sends it to the device. The input at this time is the search keyword entered by the user. The device uses the built-in camera and microphone to analyze the user's facial expressions and tone of voice to collect emotional data. The emotional data is analyzed using Microsoft Azure's emotion recognition API. This results in the search keyword and the user's emotional data.

[0607] Step 3:

[0608] Submitting a search request

[0609] The device combines the acquired search keywords and emotion data to construct a search request. The input at this time is the search keywords and emotion data. The device uses JavaScript to form the request in JSON format and sends it to the server as an HTTP POST request. This results in a search request to the server.

[0610] Step 4:

[0611] Retrieving relevant documents and using sentiment data

[0612] The server analyzes the received search request and extracts search keywords and sentiment data. The input is the search request, and the output is the search keywords and sentiment data. The server searches the index data store and extracts the most relevant documents based on the search keywords. A generative AI model (e.g., OpenAI's GPT-3) is used for this data extraction. The generative AI model takes sentiment data into account to generate optimal search results, resulting in the most relevant search results.

[0613] Step 5:

[0614] Displaying search results

[0615] The server sends the generated search results to the terminal. The input at this time is the generated search results, and the output is the search result response from the server. The terminal renders the received search results in HTML format and displays them to the user. React.js is used for display. This allows the user to quickly access the desired information by referring to the titles and access paths of related documents.

[0616] Step 6:

[0617] Recording and learning search history and emotion data

[0618] The server records the user's search history and sentiment data in a database (e.g., Apache Hadoop). The input is the user's search history and sentiment data. The output is the search history and sentiment data after it has been saved in the database. This data is used to improve the accuracy of search results from the next time onwards using a machine learning model (e.g., Scikit-learn). This updates the user profile, and more relevant results will be provided for the next search.

[0619] Step 7:

[0620] Using generative AI models

[0621] The server uses a generative AI model to understand the context of the user's search keywords and emotional data. The inputs are the search keywords and emotional data. The output is search results optimized based on the context. The generative AI model (e.g., OpenAI's GPT-3) generates search results that also take emotional data into account. This provides search results that are appropriate for the user's emotions.

[0622] (Application example 2)

[0623] 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."

[0624] Conventional information search systems have difficulty quickly and appropriately providing the necessary information from a large amount of information. Furthermore, they lacked personalization functionality that took into account the user's search trends and emotional state, making it impossible to provide the most useful information for the user. As a result, information search efficiency declined, and user satisfaction was compromised.

[0625] 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 indexing necessary information from a large amount of information, means for receiving search keywords from the user, means for extracting related documents from the indexed data based on the search keywords, means for displaying search results to the user, means for recording the user's search history and emotional data and learning the user's search tendencies and emotional state, and means for personalizing search results based on the user's emotional data. This makes it possible to quickly and accurately provide information related to the user's search keywords and emotional data.

[0626] "A means of indexing necessary information from a large amount of information" is the process by which a server retrieves all document information from a content management system and stores it in an index data store to speed up searches.

[0627] The "means for receiving search keywords from a user" refers to a process in which a user inputs search keywords to search for specific information through a terminal.

[0628] "Means for extracting relevant documents from indexed data based on search keywords" refers to the process by which the server searches the index based on the search keywords and extracts the most relevant documents.

[0629] The "means for displaying search results to the user" refers to the process by which the terminal displays the search results received from the server on the screen, allowing the user to access the desired information.

[0630] "Means for recording users' search history and emotional data and learning their search tendencies and emotional states" refers to the process by which the server records users' past search history and emotional data, and updates the algorithm based on this data to understand users' search tendencies and emotions.

[0631] "Means for personalizing search results based on user emotional data" refers to a process in which the server utilizes the user's emotional data obtained from the emotion engine to optimize search results to adapt to the user's emotional state.

[0632] "Means for updating the machine learning model based on search query input and sentiment data to reflect the user's tendencies and emotional state for future searches" refers to the process by which the server uses the user's search query and sentiment data to train the machine learning model to provide results that adapt to the user's individual tendencies and emotional state for future searches.

[0633] "Means of using a generative AI model to quickly and accurately provide information related to a user's search keywords and sentiment data" refers to the process by which a generative AI model understands a user's search keywords and sentiment data and generates optimal search results.

[0634] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends and emotions. This system works in cooperation with a server, a terminal, and an emotion engine.

[0635] The server first retrieves all document information from the content management system and indexes them. The indexed data is used to speed up searches and is stored in an index data store. Indexing occurs automatically whenever a new document is added.

[0636] To search for specific information, a user inputs search keywords into the device. The device also has an emotion engine that recognizes the user's emotions, and collects the user's emotion data.

[0637] The terminal receives search keywords and emotion data input by the user and transmits this information to the server.

[0638] The server searches the index based on the received search keywords to extract the most relevant documents, taking into account the user's emotional data and optimizing the search results in a format that is appropriate for the user's emotions.

[0639] The terminal displays the search results received from the server to the user. The displayed search results include titles and paths, which the user can refer to to access the desired information.

[0640] The server records the user's search history and emotion data, associates them with the user profile, and stores them. Furthermore, by updating the machine learning model, the server can provide more relevant search results based on the user's search habits and emotions the next time the user searches.

[0641] The server uses a generative AI model to generate optimal search results based on the user's search keywords and emotional data. Specifically, the generative AI model can generate search results taking into account the user's emotional data.

[0642] For example, consider a case where a user searches for "soothing music." The user enters "soothing music" into the device, and the emotion engine detects "stress." The device sends the search keywords and emotion data to the server. The server extracts documents related to "soothing music" from the indexed data and provides a list of music that helps reduce stress. In this case, the generative AI model uses prompts such as the following to generate optimal search results:

[0643] Example prompt sentence:

[0644] "A user searches for 'soothing music'. The user's emotional state is 'stressed'. Generate a list of music that will help reduce stress."

[0645] This makes it possible to provide information quickly and accurately according to the user's emotional state and search tendencies, thereby realizing a system that contributes to improving user satisfaction.

[0646] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0647] Step 1:

[0648] The server retrieves all document information from the content management system and indexes it. This indexed data is used for faster searches and is stored in an index data store. The input is the document data from the content management system and the output is the indexed data.

[0649] Step 2:

[0650] To search for specific information, a user inputs search keywords into the device. The device also incorporates an emotion engine that recognizes the user's emotions, and simultaneously collects the user's emotional data. The input is the user's search keywords and emotional data, and the output is the integrated information of these data.

[0651] Step 3:

[0652] The terminal receives search keywords and emotion data entered by the user and transmits this information to the server. The input is the search keywords and emotion data, and the output is the data transmission result to the server.

[0653] Step 4:

[0654] The server searches the index data store based on the received search keywords to extract the most relevant documents, taking into account the user's emotional data and optimizing the search results in a format that is appropriate for the user's emotions. The input is the search keywords and emotional data, and the output is personalized search results.

[0655] Step 5:

[0656] The terminal displays the search results received from the server to the user. The displayed search results include a title and a path, which the user can refer to to access the desired information. The input is the search results from the server, and the output is the search results displayed to the user.

[0657] Step 6:

[0658] The server records the user's search history and emotion data, associates them with the user profile, and saves them. Furthermore, by updating the machine learning model, the server can provide more appropriate search results based on the user's search trends and emotions the next time they search. The input is the search history and emotion data, and the output is the updated machine learning model.

[0659] Step 7:

[0660] The server uses a generative AI model to generate optimal search results based on the user's search keywords and emotional data. Specifically, the generative AI model can generate search results taking into account the user's emotional data. The input is the search keywords, emotional data, and the generative AI model's prompt, and the output is personalized search results.

[0661] For example, if a user searches for "soothing music," an example prompt might look like this:

[0662] "A user searches for 'soothing music'. The user's emotional state is 'stressed'. Generate a list of music that will help reduce stress."

[0663] 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.

[0664] 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.

[0665] 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.

[0666] [Third embodiment]

[0667] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0668] 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.

[0669] 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).

[0670] 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.

[0671] 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.

[0672] 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).

[0673] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0674] 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.

[0675] 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.

[0676] 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.

[0677] 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.

[0678] 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."

[0679] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends. This system works in cooperation with a server, a terminal, and a user.

[0680] Data Indexing

[0681] The server first retrieves all document information from the content management system and indexes them. The indexed data is used to speed up searches. Indexing occurs automatically whenever a new document is added.

[0682] Enter search keywords

[0683] A user inputs search keywords into a terminal to search for specific information, and the search keywords are transmitted to a server via the terminal.

[0684] Find related documents

[0685] The server searches the index based on the received search keywords to extract the most relevant documents, using rapid references to the indexed data to generate optimal search results.

[0686] Displaying search results

[0687] The terminal displays the search results received from the server to the user. The displayed search results include titles and paths, which the user can refer to to access the desired information.

[0688] Record and learn from your search history

[0689] The server records the user's search history and updates the machine learning model based on this information, allowing the user's search trends to be reflected in the next search, resulting in more relevant search results.

[0690] Using generative AI models

[0691] Additionally, the server uses a generative AI model to understand the context of the search keywords entered and generate optimal search results, which goes beyond simple keyword matching to achieve more advanced searches.

[0692] Specific examples

[0693] For example, consider a case where a user searches for information about "project management." The user types "project management" into a terminal. The terminal sends this input to a server. The server searches the index based on the keyword "project management" and extracts relevant documents.

[0694] As a search result, the server extracts documents with titles such as "Project Management Fundamentals" and "Efficient Project Management Methods" and sends them to the terminal. The terminal displays them to the user, allowing the user to quickly access the information they need.

[0695] The server also records the user's search history and learns to prioritize displaying information related to "project management" in subsequent searches.

[0696] By combining users' search history with generative AI models, this system can provide more accurate and personalized search results, significantly reducing the time required to search for information.

[0697] In this way, Knowledge Comet provides users with a fast and accurate information search experience, contributing to improved overall business efficiency and productivity.

[0698] The processing flow will be explained below.

[0699] Step 1:

[0700] The server retrieves all document information from the content management system, indexes these documents, transforms the indexed data into a structure that facilitates fast searches, and stores it in an index data store.

[0701] Step 2:

[0702] A user inputs search keywords into a terminal to search for specific information. The search keywords input by a user are specific words or phrases.

[0703] Step 3:

[0704] The terminal receives search keywords entered by the user and transmits this information to the server, where the search keywords are processed.

[0705] Step 4:

[0706] The server takes the received search keywords and searches for relevant documents from the index data store. The search query is analyzed by a parser to extract the most relevant documents.

[0707] Step 5:

[0708] The server compiles the extracted related document information (e.g., title, path) as search results, and formats the search results in a user-friendly format.

[0709] Step 6:

[0710] The server sends the formatted search results to the terminal, which include the document title and path.

[0711] Step 7:

[0712] The terminal displays the search results received from the server to the user, who can then refer to the displayed results and access the information they need.

[0713] Step 8:

[0714] The server records the user's search queries and stores them as the user's search history, which is associated with the user's profile.

[0715] Step 9:

[0716] The server uses the saved search history to update the machine learning model, which is then used to provide better search results the next time the user searches.

[0717] Step 10:

[0718] The server uses a generative AI model to understand the context of the user's search keywords, generating advanced search results that go beyond simple keyword matching.

[0719] Step 11:

[0720] Users will be able to get more relevant search results faster on subsequent searches, improving their information-seeking experience.

[0721] Example 1

[0722] 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."

[0723] Conventional information retrieval systems have difficulty quickly and accurately searching for the necessary information from a large amount of information, and it is difficult to provide personalized search results that appropriately reflect the user's search trends. Furthermore, because they rely on simple keyword matching, they are unable to realize advanced searches that understand the context. Furthermore, when a user conducts multiple searches, the system lacks the ability to fully learn the user's search history and utilize it for future searches.

[0724] 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.

[0725] In this invention, the server includes: means for indexing necessary information from a large amount of information; means for receiving search keywords from a user; means for extracting related data from the indexed data based on the search keywords; means for displaying search results to the user; means for recording the user's search history and learning search trends; and means for analyzing the context of the entered search keywords using a generative AI model to generate optimal search results. By combining the user's search history and the generative AI model, it is possible to provide more accurate and personalized search results and significantly reduce the time required for information search.

[0726] "Indexing" is the process of organizing and classifying large amounts of information based on specific criteria so that it can be easily searched.

[0727] A "search keyword" is a word or phrase that a user enters when searching for specific information.

[0728] "Related data" refers to data items related to the targeted information that are extracted based on the user's search keywords.

[0729] "Search results" are a list of related data returned in response to a user's search keywords and displayed to the user.

[0730] "Search history" is a record of searches that a user has performed up to now, and includes information such as date and time and keywords.

[0731] "Search trends" are patterns and characteristics extracted from a user's past search behavior.

[0732] A "generative AI model" is a mathematical model that uses artificial intelligence to analyze the context and meaning of input data and generate appropriate output.

[0733] "Context" is other information or data that surrounds a keyword or statement and influences its meaning and interpretation.

[0734] "Personalization" means providing services and content that are specifically tailored and optimized based on individual user characteristics and behavior.

[0735] "Information retrieval" is the process of extracting specific information from a particular database or index.

[0736] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends. This system works in cooperation with a server, a terminal, and a user.

[0737] Data Indexing

[0738] The server first retrieves all document information from the content management system and indexes it. This process uses search engine software such as ElasticSearch or Apache Lucene. Every time a new document is added, the server automatically updates the index, speeding up searches.

[0739] Enter search keywords

[0740] A user inputs keywords into the device to search for specific information. For example, if a user inputs the keyword "project management," the device sends this input to the server via a REST API. The search keywords are sent to the server in JSON format using an AJAX request.

[0741] Find related documents

[0742] The server searches the index based on the received search keywords, generates and executes an ElasticSearch search query to extract the most relevant documents, and then packages the extracted document metadata (title, path, etc.) in JSON format and sends it to the device as a response.

[0743] Displaying search results

[0744] The terminal displays the search results received from the server to the user. Specifically, JavaScript is used to dynamically generate HTML elements and create a search result list. For example, the search results may display titles such as "Project Management Basics" and "Efficient Project Management Methods," with each title serving as a link to the corresponding document's detailed page.

[0745] Record and learn from your search history

[0746] The server records the user's search history. Specifically, it stores the search keywords and search execution timestamps in MongoDB, a NoSQL database. This recorded data is used to update machine learning models (e.g., TensorFlow and Scikit-learn) and learn the user's search patterns. This allows the server to provide more personalized search results the next time the user searches.

[0747] Using generative AI models

[0748] Furthermore, the server uses a generative AI model to understand the context of the search keywords entered. For example, by using OpenAI's GPT-3 model, it achieves a more advanced search than simple keyword matching. The generative AI model understands the context and generates advanced search results. Specific prompts could be something like:

[0749] "What are some basic project management information?"

[0750] "I'd like to get an overview of efficient project management methods."

[0751] Specific examples

[0752] For example, a user may search for information related to "project management." The user types "project management" into their device, which then sends this input to the server via an AJAX request. The server uses ElasticSearch to extract relevant documents from the index and returns the document information to the device. The device displays the search results to the user, who can click links to access more information. The server then records this search history and learns to prioritize displaying information related to "project management" in future searches.

[0753] By combining users' search history with generative AI models, the system can provide more accurate and personalized search results, significantly reducing the time it takes to find information.

[0754] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0755] Step 1: Index the data

[0756] The server first retrieves all document information from the content management system. This is done using SQL queries or API requests. To index the retrieved document information, ElasticSearch is used to organize and classify the data and create an index that enables fast searches. This process is automatically triggered each time a new document is added, and the index is updated. Specifically, when a new document is added, the server automatically analyzes the document and rebuilds the appropriate index.

[0757] Input: Document information retrieved from the content management system

[0758] Output: Data indexed by ElasticSearch

[0759] Step 2: Enter search keywords

[0760] A user enters keywords into a device to search for specific information. For example, they enter "project management" into the search bar of a web browser. The entered search keywords are sent to the server in JSON format via an AJAX request. Specifically, the device receives the user input, generates a request including the keywords, and sends it to the server.

[0761] Input: The search keyword entered by the user

[0762] Output: Search keyword request in JSON format

[0763] Step 3: Finding relevant documentation

[0764] The server searches the index based on the received search keywords. It generates and executes an ElasticSearch search query to extract the most relevant documents. The extracted document metadata (title, path, etc.) is packaged in JSON format and sent to the device as a response. Specifically, the server generates a search query and queries the indexed data to select relevant documents.

[0765] Input: Search keyword request in JSON format

[0766] Output: Associated document metadata in JSON format

[0767] Step 4: Viewing search results

[0768] The terminal displays the search results received from the server to the user. JavaScript is used to dynamically generate HTML elements and create a search result list. For example, a list of titles such as "Project Management Basics" and "Efficient Project Management Methods" is displayed, with each item in the list providing a link to a detailed page. Specifically, the terminal parses the received JSON data, generates the necessary HTML structure, and inserts it into the page.

[0769] Input: Associated document metadata in JSON format

[0770] Output: A list of search results that the user can see in their browser

[0771] Step 5: Record and learn from your search history

[0772] The server records the user's search history. Specifically, it saves search keywords and search execution timestamps in a NoSQL database such as MongoDB. This data is analyzed by a machine learning model (TensorFlow or Scikit-learn) to learn the user's search patterns. This makes it possible to provide more personalized results the next time the user searches. Specifically, the server retrieves the recorded search history and uses a machine learning algorithm to generate a training dataset and update the model.

[0773] Input: Search keyword and timestamp

[0774] Output: Machine learning model updates

[0775] Step 6: Use the generative AI model

[0776] The server uses a generative AI model to understand the context of the entered search keywords. For example, it uses OpenAI's GPT-3 model to generate context-analyzing prompts. The generative AI model analyzes the meaning of the keywords and generates advanced search results. Specifically, the server sends prompts to the GPT-3 API, analyzes the results, and filters and re-evaluates relevant documents.

[0777] Input: Search keywords and prompt text

[0778] Output: Contextualized, advanced search results

[0779] The above is the specific processing flow of this system, explaining what kind of data processing and data calculation is performed at each step and what kind of output is obtained as a result.

[0780] (Application example 1)

[0781] 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."

[0782] Current information search systems in content distribution services have difficulty quickly and accurately searching for the necessary information from a large amount of information and providing personalized search results based on the user's search history. Another problem is the lack of advanced search functions that utilize generative AI models and functions that learn the user's search trends to optimize search results for future searches.

[0783] 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.

[0784] In this invention, the server includes a means for indexing necessary information from a large amount of information, a means for receiving search keywords from a user, and a means for extracting related documents from the indexed data based on the search keywords. This adds a function for recommending personalized content using the user's search history and a machine learning model, enabling fast and accurate information retrieval. Furthermore, the generative AI model can be used to understand context and generate more appropriate search results, improving user convenience.

[0785] text

[0786] "A large amount of information" refers to a large number of documents and data, indicating a huge amount of information.

[0787] "Indexing" is the process of pre-organizing information to make searching more efficient and providing quick access to key data.

[0788] "Search keywords" refer to specific words or phrases that users enter to search for information.

[0789] "Related documents" refer to documents with related content that are extracted based on the search keywords.

[0790] "User search history" refers to records of past search queries and their results.

[0791] "Search trends" refer to search patterns and preferences derived from a user's past search history.

[0792] A "machine learning model" refers to an algorithm that learns from large amounts of data and makes inferences and predictions.

[0793] A "generative AI model" is an artificial intelligence model that understands the context of the search keywords entered and generates optimal results based on that.

[0794] A "content distribution service" refers to a service that provides users with various content such as videos, articles, and music via the Internet.

[0795] "Personalized content" refers to content that is individually tailored to a user based on their preferences and past behavior.

[0796] "Recommendation methods" refer to methods of selecting and presenting appropriate content based on a user's search history and trends.

[0797]

[0798] This invention is a system for quickly and accurately searching for necessary information from a large amount of information and providing personalized search results based on the user's search history. In particular, this system enhances the personalized content recommendation function in content distribution services by utilizing the user's search history and machine learning models.

[0799] Hardware and Software Configuration

[0800] 1. Server: The server works with multiple databases to index and store large amounts of documents. The indexed data is managed using a search engine such as Elasticsearch.

[0801] 2. User device: User devices such as smartphones and head-mounted displays (HMDs) receive search keywords from users and send them to the server. They also display search results and recommended content to users.

[0802] 3. Generative AI Model: The server uses a generative AI model (e.g., Hugging Face Transformers) to understand the context of the search keywords and generate optimal search results based on that. This generative AI model is also used to update the search history and personalized models.

[0803] System Operation

[0804] 1. Data Indexing:

[0805] The server retrieves large amounts of document information and indexes it using Elasticsearch, which is automatically updated whenever new documents are added.

[0806] 2. Receiving search keywords:

[0807] Users input search keywords via their smartphone or HMD, which are then sent to the server.

[0808] 3. Search for related documents:

[0809] The server quickly extracts relevant documents from the indexed data based on search keywords, and the extracted documents are personalized based on the user's search history.

[0810] 4. Displaying search results:

[0811] The user terminal displays the received search results to the user, including document titles and links that the user can refer to to access more detailed information.

[0812] 5. Learn your search history:

[0813] The server records the user's search history and updates the machine learning model based on this information, allowing the user's search trends to be reflected in subsequent searches, resulting in more relevant search results.

[0814] 6. Personalized Recommendations:

[0815] The server uses a generative AI model to recommend optimal content based on the user's search history and behavioral patterns, allowing users to quickly access new content based on their past search history and behavior.

[0816] Examples and prompts

[0817] A specific example is when a user searches for "the latest trends in project management" on a smartphone. In this case, the user types "the latest trends in project management."

[0818] Examples of prompts:

[0819] "Latest Trends in Project Management"

[0820] This prompt is sent to the server, which then extracts relevant documents from the indexed data and sends them to the user's device. The device displays them to the user, allowing them to quickly access the information they need. This search history is also recorded and reflected in future searches.

[0821] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0822] Step 1:

[0823] The server retrieves large amounts of document information. Specifically, it retrieves document data from a content management system and indexes it using a search engine such as Elasticsearch. It receives document information as input and outputs it as indexed data. This process is performed to reduce search time for search queries.

[0824] Step 2:

[0825] The user inputs search keywords through a smartphone or head-mounted display (HMD). The input keywords are sent from the user device to the server. In this step, the input information about what the user is looking for is collected and becomes input data for the next processing step.

[0826] Step 3:

[0827] The server searches the indexed data based on the received search keywords, using Elasticsearch to quickly extract documents related to the search keywords. The input is the search keywords submitted by the user and the indexed data, and the output is a list of relevant documents.

[0828] Step 4:

[0829] The server uses a generative AI model to understand the context of the search keywords and generate optimal search results based on that. Specifically, it uses a generative AI model, such as Hugging Face's Transformers, to evaluate related documents and select the best results. In this step, the search keywords and related documents are used as input, and personalized search results are generated as output.

[0830] Step 5:

[0831] The server sends the generated search results to the user terminal. The user terminal displays the received search results to the user. The displayed information includes document titles and links, allowing the user to quickly access the information they need. The input of this step is the generated search results, and the output is the search results displayed to the user.

[0832] Step 6:

[0833] The server records the user's search history, specifically the search queries and documents selected by the user, and stores them in a database. The input is the user's history of actions, and the output is a stored search history. This data is used to optimize future searches.

[0834] Step 7:

[0835] The server updates the machine learning model based on the search history. Specifically, it uses the saved search history to train the model to reflect the user's preferences in the next search. The input for this step is the user's search history, and the output is an updated machine learning model.

[0836] Step 8:

[0837] The server then uses the updated machine learning model to recommend personalized content based on the user's search history and behavioral patterns. In this step, the user's search history and the machine learning model are used as inputs, and a personalized content list is generated as output.

[0838] As described above, the system goes through a series of processing steps to provide users with a fast and personalized information retrieval experience.

[0839] 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.

[0840] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends and emotions. This system works in cooperation with a server, a terminal, and a user.

[0841] Data Indexing

[0842] The server first retrieves all document information from the content management system and indexes them. The indexed data is used to speed up searches and is stored in an index data store. Indexing occurs automatically whenever a new document is added.

[0843] Enter search keywords and sentiment data

[0844] Users input search keywords into their devices to search for specific information. The devices also have an emotion engine that recognizes the user's emotions, and collects the user's emotion data.

[0845] Submitting a search request

[0846] The terminal receives search keywords and emotion data input by the user and transmits this information to the server.

[0847] Retrieving relevant documents and using sentiment data

[0848] The server searches the index based on the received search keywords to extract the most relevant documents, taking into account the user's emotional data provided by the emotion engine and optimizing the search results in a format appropriate to the user's emotions.

[0849] Displaying search results

[0850] The terminal displays the search results received from the server to the user. The displayed search results include titles and paths, which the user can refer to to access the desired information.

[0851] Recording and learning search history and emotion data

[0852] The server records the user's search history and emotion data, associates them with the user profile, and stores them. Furthermore, by updating the machine learning model, the server can provide more appropriate search results based on the user's search habits and emotions the next time the user searches.

[0853] Using generative AI models

[0854] The server uses a generative AI model to understand the context of the user's search keywords and generate optimal search results, which also take into account sentiment data.

[0855] Specific examples

[0856] For example, consider a case where a user searches for information about "project management." The user types "project management" into the device. At the same time, the emotion engine detects that the user is feeling stressed. The device sends the search keywords and emotion data to the server. The server searches the index based on the keyword "project management" and extracts relevant documents.

[0857] As a search result, the server extracts documents with titles such as "Project Management Fundamentals" and "Efficient Project Management Methods" and sends them to the terminal. The terminal displays them to the user, allowing the user to quickly access the information they need.

[0858] Additionally, the server records the user's search history and emotional data, and learns to prioritize information related to "project management" and their emotional state in future searches. If the user is feeling stressed, the server may also display information on relaxation techniques and articles that can help reduce stress.

[0859] This system combines a user's search history with emotional data to provide more relevant information that is in line with the user's emotions quickly and accurately, improving the information search experience and ultimately increasing work efficiency and productivity.

[0860] The processing flow will be explained below.

[0861] Step 1:

[0862] The server retrieves all document information from the content management system and indexes it. The indexed data is stored in an index data store for fast searching. This index is automatically updated whenever new documents are added.

[0863] Step 2:

[0864] A user inputs search keywords into a device to search for specific information. The device, which is equipped with an emotion engine, simultaneously recognizes the user's emotional state using technologies such as voice analysis and facial expression recognition.

[0865] Step 3:

[0866] The terminal receives search keywords and emotion data from the user and transmits this information to the server in a transmission format that includes the search keywords and emotion data.

[0867] Step 4:

[0868] The server searches for relevant documents from the index data store based on the received search keywords. The search query is analyzed by a parser to extract the most relevant documents. At the same time, the search results are optimized based on sentiment data.

[0869] Step 5:

[0870] The server then aggregates the extracted relevant document information into search results that take into account the user's emotional state and are formatted to best fit the user's current emotions.

[0871] Step 6:

[0872] The server sends the formatted search results to the terminal, which include the document title and path.

[0873] Step 7:

[0874] The terminal displays the search results received from the server to the user, who can then refer to the displayed results and access the information they need.

[0875] Step 8:

[0876] The server records the user's search queries and sentiment data, which are then associated with the user's profile to aid in future searches.

[0877] Step 9:

[0878] The server uses the recorded search history and sentiment data to update the machine learning model, which is then used to provide more accurate search results the next time the user searches.

[0879] Step 10:

[0880] The server uses a generative AI model to understand the context of the user's search keywords and generate optimal search results, which also take into account sentiment data.

[0881] Step 11:

[0882] Users will get faster, more relevant search results on subsequent searches, improving their information-seeking experience and overall business efficiency.

[0883] Specific examples

[0884] For example, if a user searches for information about "project management," the user enters "project management" into the device. The emotion engine detects stress from the user's tone of voice. The device then sends the search keyword "project management" and emotion data (stress state) to the server.

[0885] The server searches the index data store based on the received search keyword "project management" and extracts relevant documents. In addition, taking into account the emotional data indicating that the user is feeling stressed, documents related to stress management and relaxation methods are also included in the search results.

[0886] As search results, the server extracts documents with titles such as "Project Management Basics" and "Efficient Project Management Methods," as well as documents such as "Methods for managing stress in project management," and sends these to the terminal.

[0887] The device displays the results to the user, allowing them to quickly and appropriately access the information they need. The server records this search history and sentiment data and updates the machine learning model to help with future searches.

[0888] In this way, Knowledge Comet combines users' search history and sentiment data to provide more relevant and contextually appropriate information quickly and accurately, resulting in an improved information search experience and increased work efficiency and productivity.

[0889] Example 2

[0890] 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."

[0891] In today's world, where a vast amount of information is digitized, it remains difficult for users to quickly and accurately search for the information they need. Conventional search systems only provide simple keyword-based search functions and are unable to fully consider users' emotions and search trends. As a result, users' needs cannot be fully met, resulting in a limited search experience. Furthermore, search results that are not tailored to users' emotions can cause frustration and dissatisfaction. The present invention aims to address these issues and improve the information search experience by providing personalized search results that take into account users' emotions and search history.

[0892] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0893] In this invention, the server includes: a means for indexing necessary information from a large amount of information; a means for receiving search keywords from a user; a means for analyzing the user's emotions using collected emotion data; a means for extracting related documents from the indexed data based on the search keywords and emotion data; a means for optimizing search results in a format appropriate for the user's emotions and displaying them to the user; a means for recording the user's search history and emotion data and learning search trends; and a means for generating optimal search results using a generative AI model. This enables the provision of search results that take into account not only the user's search keywords but also the user's emotional state, resulting in a more satisfying information search experience for the user. Furthermore, continuous learning using the machine learning model can improve the accuracy of search results from the next search onwards.

[0894] "Indexing" is the process of organizing data and information according to specific rules and attaching tags and keys to make searching more efficient.

[0895] "Search keywords" are words or phrases that a user enters to search for specific information.

[0896] "Emotion data" refers to information that expresses the user's emotional state using numerical values ​​or categories, and includes, for example, joy, sadness, surprise, and the like.

[0897] "Related documents" are documents or information that are determined to be most relevant based on the search keywords and emotion data entered by the user.

[0898] "Optimization" means adjusting something to the most effective or efficient state for specific conditions or goals.

[0899] "Search history" refers to a record of a user's past search queries, links they have clicked, and so on.

[0900] A "generative AI model" is artificial intelligence that uses advanced algorithms, such as deep learning, to generate text and other forms of data.

[0901] A "machine learning model" is an algorithm or statistical model that is trained on data to automatically perform a specific task.

[0902] A "user profile" is a data set that aggregates a user's basic information, search history, emotional data, etc.

[0903] "Natural language processing" is a technology that allows machines to understand, analyze, and generate human language.

[0904] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends and emotions. This system works in cooperation with a server, a terminal, and a user.

[0905] Data Indexing

[0906] The server first retrieves all document information from the content management system and indexes it. Indexing is done using the Apache Lucene search engine, and the results are stored in MongoDB. The indexed data is used to speed up searches and is automatically updated whenever new documents are added.

[0907] Enter search keywords and sentiment data

[0908] A user inputs search keywords into a device to search for specific information. The device is equipped with a camera that analyzes the user's facial expressions and a microphone that analyzes the tone of the user's voice, and collects the user's emotional data using an emotion recognition API (for example, Microsoft Azure's emotion recognition API).

[0909] Submitting a search request

[0910] The device receives search keywords and emotion data entered by the user and sends this information to the server. The request is constructed in JSON format using JavaScript and sent to the server as an HTTP POST request.

[0911] Retrieving relevant documents and using sentiment data

[0912] The server analyzes the received search keywords and sentiment data, searches the index data store, extracts the most relevant documents, and feeds the results into a generative AI model (e.g., OpenAI's GPT-3) to generate search results that fit the user's sentiment.

[0913] Displaying search results

[0914] The terminal renders the search results received from the server in HTML format and displays them in the browser. The search results include document titles and access paths, which users can refer to to access the desired information. The terminal uses React.js.

[0915] Recording and learning search history and emotion data

[0916] The server records the user's search history and sentiment data in a database (e.g., Apache Hadoop) and continuously updates a machine learning model (e.g., Scikit-learn). This data is associated with the user's profile and used to provide better search results the next time the user searches.

[0917] Examples and prompts

[0918] As a concrete example, consider a case where a user searches for information on "project management." In this case, the user enters "project management" into the device, and the emotion engine detects that the user is feeling stressed. The device sends the search keywords and emotion data to the server, which then extracts related documents based on the keyword "project management." In this case, a generative AI model is used to generate search results, and documents with titles such as "Project Management Basics" and "Efficient Project Management Methods" are sent to the device. The device displays these results to the user, allowing them to quickly access the information they need.

[0919] Example prompt sentence:

[0920] "In a search system that is integrated with an emotion engine, if a user searches for the keyword 'project management' and detects that they are feeling stressed, what search results should be displayed? Use a generative AI model to suggest the most appropriate search results."

[0921] This system combines a user's search history with emotional data to provide more relevant information that is in line with the user's emotions quickly and accurately, improving the information search experience. Furthermore, continuous learning using machine learning models improves the accuracy of search results from the next time onwards.

[0922] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0923] Step 1:

[0924] Data Indexing

[0925] The server retrieves all document information from the content management system. The input is the document information from the content management system. The server parses and indexes the document information using Apache Lucene. The indexed data is stored in MongoDB, providing an index data store for efficient data searches. When new documents are added, the index is automatically updated.

[0926] Step 2:

[0927] Enter search keywords and sentiment data

[0928] The user enters a search keyword (e.g., "project management") into the search bar and sends it to the device. The input at this time is the search keyword entered by the user. The device uses the built-in camera and microphone to analyze the user's facial expressions and tone of voice to collect emotional data. The emotional data is analyzed using Microsoft Azure's emotion recognition API. This results in the search keyword and the user's emotional data.

[0929] Step 3:

[0930] Submitting a search request

[0931] The device combines the acquired search keywords and emotion data to construct a search request. The input at this time is the search keywords and emotion data. The device uses JavaScript to form the request in JSON format and sends it to the server as an HTTP POST request. This results in a search request to the server.

[0932] Step 4:

[0933] Retrieving relevant documents and using sentiment data

[0934] The server analyzes the received search request and extracts search keywords and sentiment data. The input is the search request, and the output is the search keywords and sentiment data. The server searches the index data store and extracts the most relevant documents based on the search keywords. A generative AI model (e.g., OpenAI's GPT-3) is used for this data extraction. The generative AI model takes sentiment data into account to generate optimal search results, resulting in the most relevant search results.

[0935] Step 5:

[0936] Displaying search results

[0937] The server sends the generated search results to the terminal. The input at this time is the generated search results, and the output is the search result response from the server. The terminal renders the received search results in HTML format and displays them to the user. React.js is used for display. This allows the user to quickly access the desired information by referring to the titles and access paths of related documents.

[0938] Step 6:

[0939] Recording and learning search history and emotion data

[0940] The server records the user's search history and sentiment data in a database (e.g., Apache Hadoop). The input is the user's search history and sentiment data. The output is the search history and sentiment data after it has been saved in the database. This data is used to improve the accuracy of search results from the next time onwards using a machine learning model (e.g., Scikit-learn). This updates the user profile, and more relevant results will be provided for the next search.

[0941] Step 7:

[0942] Using generative AI models

[0943] The server uses a generative AI model to understand the context of the user's search keywords and emotional data. The inputs are the search keywords and emotional data. The output is search results optimized based on the context. The generative AI model (e.g., OpenAI's GPT-3) generates search results that also take emotional data into account. This provides search results that are appropriate for the user's emotions.

[0944] (Application example 2)

[0945] 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."

[0946] Conventional information search systems have difficulty quickly and appropriately providing the necessary information from a large amount of information. Furthermore, they lacked personalization functionality that took into account the user's search trends and emotional state, making it impossible to provide the most useful information for the user. As a result, information search efficiency declined, and user satisfaction was compromised.

[0947] 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 indexing necessary information from a large amount of information, means for receiving search keywords from the user, means for extracting related documents from the indexed data based on the search keywords, means for displaying search results to the user, means for recording the user's search history and emotional data and learning the user's search tendencies and emotional state, and means for personalizing search results based on the user's emotional data. This makes it possible to quickly and accurately provide information related to the user's search keywords and emotional data.

[0948] "A means of indexing necessary information from a large amount of information" is the process by which a server retrieves all document information from a content management system and stores it in an index data store to speed up searches.

[0949] The "means for receiving search keywords from a user" refers to a process in which a user inputs search keywords to search for specific information through a terminal.

[0950] "Means for extracting relevant documents from indexed data based on search keywords" refers to the process by which the server searches the index based on the search keywords and extracts the most relevant documents.

[0951] The "means for displaying search results to the user" refers to the process by which the terminal displays the search results received from the server on the screen, allowing the user to access the desired information.

[0952] "Means for recording users' search history and emotional data and learning their search tendencies and emotional states" refers to the process by which the server records users' past search history and emotional data, and updates the algorithm based on this data to understand users' search tendencies and emotions.

[0953] "Means for personalizing search results based on user emotional data" refers to a process in which the server utilizes the user's emotional data obtained from the emotion engine to optimize search results to adapt to the user's emotional state.

[0954] "Means for updating the machine learning model based on search query input and sentiment data to reflect the user's tendencies and emotional state for future searches" refers to the process by which the server uses the user's search query and sentiment data to train the machine learning model to provide results that adapt to the user's individual tendencies and emotional state for future searches.

[0955] "Means of using a generative AI model to quickly and accurately provide information related to a user's search keywords and sentiment data" refers to the process by which a generative AI model understands a user's search keywords and sentiment data and generates optimal search results.

[0956] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends and emotions. This system works in cooperation with a server, a terminal, and an emotion engine.

[0957] The server first retrieves all document information from the content management system and indexes them. The indexed data is used to speed up searches and is stored in an index data store. Indexing occurs automatically whenever a new document is added.

[0958] To search for specific information, a user inputs search keywords into the device. The device also has an emotion engine that recognizes the user's emotions, and collects the user's emotion data.

[0959] The terminal receives search keywords and emotion data input by the user and transmits this information to the server.

[0960] The server searches the index based on the received search keywords to extract the most relevant documents, taking into account the user's emotional data and optimizing the search results in a format that is appropriate for the user's emotions.

[0961] The terminal displays the search results received from the server to the user. The displayed search results include titles and paths, which the user can refer to to access the desired information.

[0962] The server records the user's search history and emotion data, associates them with the user profile, and stores them. Furthermore, by updating the machine learning model, the server can provide more relevant search results based on the user's search habits and emotions the next time the user searches.

[0963] The server uses a generative AI model to generate optimal search results based on the user's search keywords and emotional data. Specifically, the generative AI model can generate search results taking into account the user's emotional data.

[0964] For example, consider a case where a user searches for "soothing music." The user enters "soothing music" into the device, and the emotion engine detects "stress." The device sends the search keywords and emotion data to the server. The server extracts documents related to "soothing music" from the indexed data and provides a list of music that helps reduce stress. In this case, the generative AI model uses prompts such as the following to generate optimal search results:

[0965] Example prompt sentence:

[0966] "A user searches for 'soothing music'. The user's emotional state is 'stressed'. Generate a list of music that will help reduce stress."

[0967] This makes it possible to provide information quickly and accurately according to the user's emotional state and search tendencies, thereby realizing a system that contributes to improving user satisfaction.

[0968] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0969] Step 1:

[0970] The server retrieves all document information from the content management system and indexes it. This indexed data is used for faster searches and is stored in an index data store. The input is the document data from the content management system and the output is the indexed data.

[0971] Step 2:

[0972] To search for specific information, a user inputs search keywords into the device. The device also incorporates an emotion engine that recognizes the user's emotions, and simultaneously collects the user's emotional data. The input is the user's search keywords and emotional data, and the output is the integrated information of these data.

[0973] Step 3:

[0974] The terminal receives search keywords and emotion data entered by the user and transmits this information to the server. The input is the search keywords and emotion data, and the output is the data transmission result to the server.

[0975] Step 4:

[0976] The server searches the index data store based on the received search keywords to extract the most relevant documents, taking into account the user's emotional data and optimizing the search results in a format that is appropriate for the user's emotions. The input is the search keywords and emotional data, and the output is personalized search results.

[0977] Step 5:

[0978] The terminal displays the search results received from the server to the user. The displayed search results include a title and a path, which the user can refer to to access the desired information. The input is the search results from the server, and the output is the search results displayed to the user.

[0979] Step 6:

[0980] The server records the user's search history and emotion data, associates them with the user profile, and saves them. Furthermore, by updating the machine learning model, the server can provide more appropriate search results based on the user's search trends and emotions the next time they search. The input is the search history and emotion data, and the output is the updated machine learning model.

[0981] Step 7:

[0982] The server uses a generative AI model to generate optimal search results based on the user's search keywords and emotional data. Specifically, the generative AI model can generate search results taking into account the user's emotional data. The input is the search keywords, emotional data, and the generative AI model's prompt, and the output is personalized search results.

[0983] For example, if a user searches for "soothing music," an example prompt might look like this:

[0984] "A user searches for 'soothing music'. The user's emotional state is 'stressed'. Generate a list of music that will help reduce stress."

[0985] 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.

[0986] 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.

[0987] 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.

[0988] [Fourth embodiment]

[0989] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0990] 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.

[0991] 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).

[0992] 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.

[0993] 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.

[0994] 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).

[0995] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0996] 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.

[0997] 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.

[0998] 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.

[0999] 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.

[1000] 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.

[1001] 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."

[1002] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends. This system works in cooperation with a server, a terminal, and a user.

[1003] Data Indexing

[1004] The server first retrieves all document information from the content management system and indexes them. The indexed data is used to speed up searches. Indexing occurs automatically whenever a new document is added.

[1005] Enter search keywords

[1006] A user inputs search keywords into a terminal to search for specific information, and the search keywords are transmitted to a server via the terminal.

[1007] Find related documents

[1008] The server searches the index based on the received search keywords to extract the most relevant documents, using rapid references to the indexed data to generate optimal search results.

[1009] Displaying search results

[1010] The terminal displays the search results received from the server to the user. The displayed search results include titles and paths, which the user can refer to to access the desired information.

[1011] Record and learn from your search history

[1012] The server records the user's search history and updates the machine learning model based on this information, allowing the user's search trends to be reflected in the next search, resulting in more relevant search results.

[1013] Using generative AI models

[1014] Additionally, the server uses a generative AI model to understand the context of the search keywords entered and generate optimal search results, which goes beyond simple keyword matching to achieve more advanced searches.

[1015] Specific examples

[1016] For example, consider a case where a user searches for information about "project management." The user types "project management" into a terminal. The terminal sends this input to a server. The server searches the index based on the keyword "project management" and extracts relevant documents.

[1017] As a search result, the server extracts documents with titles such as "Project Management Fundamentals" and "Efficient Project Management Methods" and sends them to the terminal. The terminal displays them to the user, allowing the user to quickly access the information they need.

[1018] The server also records the user's search history and learns to prioritize displaying information related to "project management" in subsequent searches.

[1019] By combining users' search history with generative AI models, this system can provide more accurate and personalized search results, significantly reducing the time required to search for information.

[1020] In this way, Knowledge Comet provides users with a fast and accurate information search experience, contributing to improved overall business efficiency and productivity.

[1021] The processing flow will be explained below.

[1022] Step 1:

[1023] The server retrieves all document information from the content management system, indexes these documents, transforms the indexed data into a structure that facilitates fast searches, and stores it in an index data store.

[1024] Step 2:

[1025] A user inputs search keywords into a terminal to search for specific information. The search keywords input by a user are specific words or phrases.

[1026] Step 3:

[1027] The terminal receives search keywords entered by the user and transmits this information to the server, where the search keywords are processed.

[1028] Step 4:

[1029] The server takes the received search keywords and searches for relevant documents from the index data store. The search query is analyzed by a parser to extract the most relevant documents.

[1030] Step 5:

[1031] The server compiles the extracted related document information (e.g., title, path) as search results, and formats the search results in a user-friendly format.

[1032] Step 6:

[1033] The server sends the formatted search results to the terminal, which include the document title and path.

[1034] Step 7:

[1035] The terminal displays the search results received from the server to the user, who can then refer to the displayed results and access the information they need.

[1036] Step 8:

[1037] The server records the user's search queries and stores them as the user's search history, which is associated with the user's profile.

[1038] Step 9:

[1039] The server uses the saved search history to update the machine learning model, which is then used to provide better search results the next time the user searches.

[1040] Step 10:

[1041] The server uses a generative AI model to understand the context of the user's search keywords, generating advanced search results that go beyond simple keyword matching.

[1042] Step 11:

[1043] Users will be able to get more relevant search results faster on subsequent searches, improving their information-seeking experience.

[1044] Example 1

[1045] 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."

[1046] Conventional information retrieval systems have difficulty quickly and accurately searching for the necessary information from a large amount of information, and it is difficult to provide personalized search results that appropriately reflect the user's search trends. Furthermore, because they rely on simple keyword matching, they are unable to realize advanced searches that understand the context. Furthermore, when a user conducts multiple searches, the system lacks the ability to fully learn the user's search history and utilize it for future searches.

[1047] 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.

[1048] In this invention, the server includes: means for indexing necessary information from a large amount of information; means for receiving search keywords from a user; means for extracting related data from the indexed data based on the search keywords; means for displaying search results to the user; means for recording the user's search history and learning search trends; and means for analyzing the context of the entered search keywords using a generative AI model to generate optimal search results. By combining the user's search history and the generative AI model, it is possible to provide more accurate and personalized search results and significantly reduce the time required for information search.

[1049] "Indexing" is the process of organizing and classifying large amounts of information based on specific criteria so that it can be easily searched.

[1050] A "search keyword" is a word or phrase that a user enters when searching for specific information.

[1051] "Related data" refers to data items related to the targeted information that are extracted based on the user's search keywords.

[1052] "Search results" are a list of related data returned in response to a user's search keywords and displayed to the user.

[1053] "Search history" is a record of searches that a user has performed up to now, and includes information such as date and time and keywords.

[1054] "Search trends" are patterns and characteristics extracted from a user's past search behavior.

[1055] A "generative AI model" is a mathematical model that uses artificial intelligence to analyze the context and meaning of input data and generate appropriate output.

[1056] "Context" is other information or data that surrounds a keyword or statement and influences its meaning and interpretation.

[1057] "Personalization" means providing services and content that are specifically tailored and optimized based on individual user characteristics and behavior.

[1058] "Information retrieval" is the process of extracting specific information from a particular database or index.

[1059] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends. This system works in cooperation with a server, a terminal, and a user.

[1060] Data Indexing

[1061] The server first retrieves all document information from the content management system and indexes it. This process uses search engine software such as ElasticSearch or Apache Lucene. Every time a new document is added, the server automatically updates the index, speeding up searches.

[1062] Enter search keywords

[1063] A user inputs keywords into the device to search for specific information. For example, if a user inputs the keyword "project management," the device sends this input to the server via a REST API. The search keywords are sent to the server in JSON format using an AJAX request.

[1064] Find related documents

[1065] The server searches the index based on the received search keywords, generates and executes an ElasticSearch search query to extract the most relevant documents, and then packages the extracted document metadata (title, path, etc.) in JSON format and sends it to the device as a response.

[1066] Displaying search results

[1067] The terminal displays the search results received from the server to the user. Specifically, JavaScript is used to dynamically generate HTML elements and create a search result list. For example, the search results may display titles such as "Project Management Basics" and "Efficient Project Management Methods," with each title serving as a link to the corresponding document's detailed page.

[1068] Record and learn from your search history

[1069] The server records the user's search history. Specifically, it stores the search keywords and search execution timestamps in MongoDB, a NoSQL database. This recorded data is used to update machine learning models (e.g., TensorFlow and Scikit-learn) and learn the user's search patterns. This allows the server to provide more personalized search results the next time the user searches.

[1070] Using generative AI models

[1071] Furthermore, the server uses a generative AI model to understand the context of the search keywords entered. For example, by using OpenAI's GPT-3 model, it achieves a more advanced search than simple keyword matching. The generative AI model understands the context and generates advanced search results. Specific prompts could be something like:

[1072] "What are some basic project management information?"

[1073] "I'd like to get an overview of efficient project management methods."

[1074] Specific examples

[1075] For example, a user may search for information related to "project management." The user types "project management" into their device, which then sends this input to the server via an AJAX request. The server uses ElasticSearch to extract relevant documents from the index and returns the document information to the device. The device displays the search results to the user, who can click links to access more information. The server then records this search history and learns to prioritize displaying information related to "project management" in future searches.

[1076] By combining users' search history with generative AI models, the system can provide more accurate and personalized search results, significantly reducing the time it takes to find information.

[1077] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1078] Step 1: Index the data

[1079] The server first retrieves all document information from the content management system. This is done using SQL queries or API requests. To index the retrieved document information, ElasticSearch is used to organize and classify the data and create an index that enables fast searches. This process is automatically triggered each time a new document is added, and the index is updated. Specifically, when a new document is added, the server automatically analyzes the document and rebuilds the appropriate index.

[1080] Input: Document information retrieved from the content management system

[1081] Output: Data indexed by ElasticSearch

[1082] Step 2: Enter search keywords

[1083] A user enters keywords into a device to search for specific information. For example, they enter "project management" into the search bar of a web browser. The entered search keywords are sent to the server in JSON format via an AJAX request. Specifically, the device receives the user input, generates a request including the keywords, and sends it to the server.

[1084] Input: The search keyword entered by the user

[1085] Output: Search keyword request in JSON format

[1086] Step 3: Finding relevant documentation

[1087] The server searches the index based on the received search keywords. It generates and executes an ElasticSearch search query to extract the most relevant documents. The extracted document metadata (title, path, etc.) is packaged in JSON format and sent to the device as a response. Specifically, the server generates a search query and queries the indexed data to select relevant documents.

[1088] Input: Search keyword request in JSON format

[1089] Output: Associated document metadata in JSON format

[1090] Step 4: Viewing search results

[1091] The terminal displays the search results received from the server to the user. JavaScript is used to dynamically generate HTML elements and create a search result list. For example, a list of titles such as "Project Management Basics" and "Efficient Project Management Methods" is displayed, with each item in the list providing a link to a detailed page. Specifically, the terminal parses the received JSON data, generates the necessary HTML structure, and inserts it into the page.

[1092] Input: Associated document metadata in JSON format

[1093] Output: A list of search results that the user can see in their browser

[1094] Step 5: Record and learn from your search history

[1095] The server records the user's search history. Specifically, it saves search keywords and search execution timestamps in a NoSQL database such as MongoDB. This data is analyzed by a machine learning model (TensorFlow or Scikit-learn) to learn the user's search patterns. This makes it possible to provide more personalized results the next time the user searches. Specifically, the server retrieves the recorded search history and uses a machine learning algorithm to generate a training dataset and update the model.

[1096] Input: Search keyword and timestamp

[1097] Output: Machine learning model updates

[1098] Step 6: Use the generative AI model

[1099] The server uses a generative AI model to understand the context of the entered search keywords. For example, it uses OpenAI's GPT-3 model to generate context-analyzing prompts. The generative AI model analyzes the meaning of the keywords and generates advanced search results. Specifically, the server sends prompts to the GPT-3 API, analyzes the results, and filters and re-evaluates relevant documents.

[1100] Input: Search keywords and prompt text

[1101] Output: Contextualized, advanced search results

[1102] The above is the specific processing flow of this system, explaining what kind of data processing and data calculation is performed at each step and what kind of output is obtained as a result.

[1103] (Application example 1)

[1104] 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."

[1105] Current information search systems in content distribution services have difficulty quickly and accurately searching for the necessary information from a large amount of information and providing personalized search results based on the user's search history. Another problem is the lack of advanced search functions that utilize generative AI models and functions that learn the user's search trends to optimize search results for future searches.

[1106] 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.

[1107] In this invention, the server includes a means for indexing necessary information from a large amount of information, a means for receiving search keywords from a user, and a means for extracting related documents from the indexed data based on the search keywords. This adds a function for recommending personalized content using the user's search history and a machine learning model, enabling fast and accurate information retrieval. Furthermore, the generative AI model can be used to understand context and generate more appropriate search results, improving user convenience.

[1108] text

[1109] "A large amount of information" refers to a large number of documents and data, indicating a huge amount of information.

[1110] "Indexing" is the process of pre-organizing information to make searching more efficient and providing quick access to key data.

[1111] "Search keywords" refer to specific words or phrases that users enter to search for information.

[1112] "Related documents" refer to documents with related content that are extracted based on the search keywords.

[1113] "User search history" refers to records of past search queries and their results.

[1114] "Search trends" refer to search patterns and preferences derived from a user's past search history.

[1115] A "machine learning model" refers to an algorithm that learns from large amounts of data and makes inferences and predictions.

[1116] A "generative AI model" is an artificial intelligence model that understands the context of the search keywords entered and generates optimal results based on that.

[1117] A "content distribution service" refers to a service that provides users with various content such as videos, articles, and music via the Internet.

[1118] "Personalized content" refers to content that is individually tailored to a user based on their preferences and past behavior.

[1119] "Recommendation methods" refer to methods of selecting and presenting appropriate content based on a user's search history and trends.

[1120]

[1121] This invention is a system for quickly and accurately searching for necessary information from a large amount of information and providing personalized search results based on the user's search history. In particular, this system enhances the personalized content recommendation function in content distribution services by utilizing the user's search history and machine learning models.

[1122] Hardware and Software Configuration

[1123] 1. Server: The server works with multiple databases to index and store large amounts of documents. The indexed data is managed using a search engine such as Elasticsearch.

[1124] 2. User device: User devices such as smartphones and head-mounted displays (HMDs) receive search keywords from users and send them to the server. They also display search results and recommended content to users.

[1125] 3. Generative AI Model: The server uses a generative AI model (e.g., Hugging Face Transformers) to understand the context of the search keywords and generate optimal search results based on that. This generative AI model is also used to update the search history and personalized models.

[1126] System Operation

[1127] 1. Data Indexing:

[1128] The server retrieves large amounts of document information and indexes it using Elasticsearch, which is automatically updated whenever new documents are added.

[1129] 2. Receiving search keywords:

[1130] Users input search keywords via their smartphone or HMD, which are then sent to the server.

[1131] 3. Search for related documents:

[1132] The server quickly extracts relevant documents from the indexed data based on search keywords, and the extracted documents are personalized based on the user's search history.

[1133] 4. Displaying search results:

[1134] The user terminal displays the received search results to the user, including document titles and links that the user can refer to to access more detailed information.

[1135] 5. Learn your search history:

[1136] The server records the user's search history and updates the machine learning model based on this information, allowing the user's search trends to be reflected in subsequent searches, resulting in more relevant search results.

[1137] 6. Personalized Recommendations:

[1138] The server uses a generative AI model to recommend optimal content based on the user's search history and behavioral patterns, allowing users to quickly access new content based on their past search history and behavior.

[1139] Examples and prompts

[1140] A specific example is when a user searches for "the latest trends in project management" on a smartphone. In this case, the user types "the latest trends in project management."

[1141] Examples of prompts:

[1142] "Latest Trends in Project Management"

[1143] This prompt is sent to the server, which then extracts relevant documents from the indexed data and sends them to the user's device. The device displays them to the user, allowing them to quickly access the information they need. This search history is also recorded and reflected in future searches.

[1144] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1145] Step 1:

[1146] The server retrieves large amounts of document information. Specifically, it retrieves document data from a content management system and indexes it using a search engine such as Elasticsearch. It receives document information as input and outputs it as indexed data. This process is performed to reduce search time for search queries.

[1147] Step 2:

[1148] The user inputs search keywords through a smartphone or head-mounted display (HMD). The input keywords are sent from the user device to the server. In this step, the input information about what the user is looking for is collected and becomes input data for the next processing step.

[1149] Step 3:

[1150] The server searches the indexed data based on the received search keywords, using Elasticsearch to quickly extract documents related to the search keywords. The input is the search keywords submitted by the user and the indexed data, and the output is a list of relevant documents.

[1151] Step 4:

[1152] The server uses a generative AI model to understand the context of the search keywords and generate optimal search results based on that. Specifically, it uses a generative AI model, such as Hugging Face's Transformers, to evaluate related documents and select the best results. In this step, the search keywords and related documents are used as input, and personalized search results are generated as output.

[1153] Step 5:

[1154] The server sends the generated search results to the user terminal. The user terminal displays the received search results to the user. The displayed information includes document titles and links, allowing the user to quickly access the information they need. The input of this step is the generated search results, and the output is the search results displayed to the user.

[1155] Step 6:

[1156] The server records the user's search history, specifically the search queries and documents selected by the user, and stores them in a database. The input is the user's history of actions, and the output is a stored search history. This data is used to optimize future searches.

[1157] Step 7:

[1158] The server updates the machine learning model based on the search history. Specifically, it uses the saved search history to train the model to reflect the user's preferences in the next search. The input for this step is the user's search history, and the output is an updated machine learning model.

[1159] Step 8:

[1160] The server then uses the updated machine learning model to recommend personalized content based on the user's search history and behavioral patterns. In this step, the user's search history and the machine learning model are used as inputs, and a personalized content list is generated as output.

[1161] As described above, the system goes through a series of processing steps to provide users with a fast and personalized information retrieval experience.

[1162] 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.

[1163] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends and emotions. This system works in cooperation with a server, a terminal, and a user.

[1164] Data Indexing

[1165] The server first retrieves all document information from the content management system and indexes them. The indexed data is used to speed up searches and is stored in an index data store. Indexing occurs automatically whenever a new document is added.

[1166] Enter search keywords and sentiment data

[1167] Users input search keywords into their devices to search for specific information. The devices also have an emotion engine that recognizes the user's emotions, and collects the user's emotion data.

[1168] Submitting a search request

[1169] The terminal receives search keywords and emotion data input by the user and transmits this information to the server.

[1170] Retrieving relevant documents and using sentiment data

[1171] The server searches the index based on the received search keywords to extract the most relevant documents, taking into account the user's emotional data provided by the emotion engine and optimizing the search results in a format appropriate to the user's emotions.

[1172] Displaying search results

[1173] The terminal displays the search results received from the server to the user. The displayed search results include titles and paths, which the user can refer to to access the desired information.

[1174] Recording and learning search history and emotion data

[1175] The server records the user's search history and emotion data, associates them with the user profile, and stores them. Furthermore, by updating the machine learning model, the server can provide more appropriate search results based on the user's search habits and emotions the next time the user searches.

[1176] Using generative AI models

[1177] The server uses a generative AI model to understand the context of the user's search keywords and generate optimal search results, which also take into account sentiment data.

[1178] Specific examples

[1179] For example, consider a case where a user searches for information about "project management." The user types "project management" into the device. At the same time, the emotion engine detects that the user is feeling stressed. The device sends the search keywords and emotion data to the server. The server searches the index based on the keyword "project management" and extracts relevant documents.

[1180] As a search result, the server extracts documents with titles such as "Project Management Fundamentals" and "Efficient Project Management Methods" and sends them to the terminal. The terminal displays them to the user, allowing the user to quickly access the information they need.

[1181] Additionally, the server records the user's search history and emotional data, and learns to prioritize information related to "project management" and their emotional state in future searches. If the user is feeling stressed, the server may also display information on relaxation techniques and articles that can help reduce stress.

[1182] This system combines a user's search history with emotional data to provide more relevant information that is in line with the user's emotions quickly and accurately, improving the information search experience and ultimately increasing work efficiency and productivity.

[1183] The processing flow will be explained below.

[1184] Step 1:

[1185] The server retrieves all document information from the content management system and indexes it. The indexed data is stored in an index data store for fast searching. This index is automatically updated whenever new documents are added.

[1186] Step 2:

[1187] A user inputs search keywords into a device to search for specific information. The device, which is equipped with an emotion engine, simultaneously recognizes the user's emotional state using technologies such as voice analysis and facial expression recognition.

[1188] Step 3:

[1189] The terminal receives search keywords and emotion data from the user and transmits this information to the server in a transmission format that includes the search keywords and emotion data.

[1190] Step 4:

[1191] The server searches for relevant documents from the index data store based on the received search keywords. The search query is analyzed by a parser to extract the most relevant documents. At the same time, the search results are optimized based on sentiment data.

[1192] Step 5:

[1193] The server then aggregates the extracted relevant document information into search results that take into account the user's emotional state and are formatted to best fit the user's current emotions.

[1194] Step 6:

[1195] The server sends the formatted search results to the terminal, which include the document title and path.

[1196] Step 7:

[1197] The terminal displays the search results received from the server to the user, who can then refer to the displayed results and access the information they need.

[1198] Step 8:

[1199] The server records the user's search queries and sentiment data, which are then associated with the user's profile to aid in future searches.

[1200] Step 9:

[1201] The server uses the recorded search history and sentiment data to update the machine learning model, which is then used to provide more accurate search results the next time the user searches.

[1202] Step 10:

[1203] The server uses a generative AI model to understand the context of the user's search keywords and generate optimal search results, which also take into account sentiment data.

[1204] Step 11:

[1205] Users will get faster, more relevant search results on subsequent searches, improving their information-seeking experience and overall business efficiency.

[1206] Specific examples

[1207] For example, if a user searches for information about "project management," the user enters "project management" into the device. The emotion engine detects stress from the user's tone of voice. The device then sends the search keyword "project management" and emotion data (stress state) to the server.

[1208] The server searches the index data store based on the received search keyword "project management" and extracts relevant documents. In addition, taking into account the emotional data indicating that the user is feeling stressed, documents related to stress management and relaxation methods are also included in the search results.

[1209] As search results, the server extracts documents with titles such as "Project Management Basics" and "Efficient Project Management Methods," as well as documents such as "Methods for managing stress in project management," and sends these to the terminal.

[1210] The device displays the results to the user, allowing them to quickly and appropriately access the information they need. The server records this search history and sentiment data and updates the machine learning model to help with future searches.

[1211] In this way, Knowledge Comet combines users' search history and sentiment data to provide more relevant and contextually appropriate information quickly and accurately, resulting in an improved information search experience and increased work efficiency and productivity.

[1212] Example 2

[1213] 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."

[1214] In today's world, where a vast amount of information is digitized, it remains difficult for users to quickly and accurately search for the information they need. Conventional search systems only provide simple keyword-based search functions and are unable to fully consider users' emotions and search trends. As a result, users' needs cannot be fully met, resulting in a limited search experience. Furthermore, search results that are not tailored to users' emotions can cause frustration and dissatisfaction. The present invention aims to address these issues and improve the information search experience by providing personalized search results that take into account users' emotions and search history.

[1215] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1216] In this invention, the server includes: a means for indexing necessary information from a large amount of information; a means for receiving search keywords from a user; a means for analyzing the user's emotions using collected emotion data; a means for extracting related documents from the indexed data based on the search keywords and emotion data; a means for optimizing search results in a format appropriate for the user's emotions and displaying them to the user; a means for recording the user's search history and emotion data and learning search trends; and a means for generating optimal search results using a generative AI model. This enables the provision of search results that take into account not only the user's search keywords but also the user's emotional state, resulting in a more satisfying information search experience for the user. Furthermore, continuous learning using the machine learning model can improve the accuracy of search results from the next search onwards.

[1217] "Indexing" is the process of organizing data and information according to specific rules and attaching tags and keys to make searching more efficient.

[1218] "Search keywords" are words or phrases that a user enters to search for specific information.

[1219] "Emotion data" refers to information that expresses the user's emotional state using numerical values ​​or categories, and includes, for example, joy, sadness, surprise, and the like.

[1220] "Related documents" are documents or information that are determined to be most relevant based on the search keywords and emotion data entered by the user.

[1221] "Optimization" means adjusting something to the most effective or efficient state for specific conditions or goals.

[1222] "Search history" refers to a record of a user's past search queries, links they have clicked, and so on.

[1223] A "generative AI model" is artificial intelligence that uses advanced algorithms, such as deep learning, to generate text and other forms of data.

[1224] A "machine learning model" is an algorithm or statistical model that is trained on data to automatically perform a specific task.

[1225] A "user profile" is a data set that aggregates a user's basic information, search history, emotional data, etc.

[1226] "Natural language processing" is a technology that allows machines to understand, analyze, and generate human language.

[1227] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends and emotions. This system works in cooperation with a server, a terminal, and a user.

[1228] Data Indexing

[1229] The server first retrieves all document information from the content management system and indexes it. Indexing is done using the Apache Lucene search engine, and the results are stored in MongoDB. The indexed data is used to speed up searches and is automatically updated whenever new documents are added.

[1230] Enter search keywords and sentiment data

[1231] A user inputs search keywords into a device to search for specific information. The device is equipped with a camera that analyzes the user's facial expressions and a microphone that analyzes the tone of the user's voice, and collects the user's emotional data using an emotion recognition API (for example, Microsoft Azure's emotion recognition API).

[1232] Submitting a search request

[1233] The device receives search keywords and emotion data entered by the user and sends this information to the server. The request is constructed in JSON format using JavaScript and sent to the server as an HTTP POST request.

[1234] Retrieving relevant documents and using sentiment data

[1235] The server analyzes the received search keywords and sentiment data, searches the index data store, extracts the most relevant documents, and feeds the results into a generative AI model (e.g., OpenAI's GPT-3) to generate search results that fit the user's sentiment.

[1236] Displaying search results

[1237] The terminal renders the search results received from the server in HTML format and displays them in the browser. The search results include document titles and access paths, which users can refer to to access the desired information. The terminal uses React.js.

[1238] Recording and learning search history and emotion data

[1239] The server records the user's search history and sentiment data in a database (e.g., Apache Hadoop) and continuously updates a machine learning model (e.g., Scikit-learn). This data is associated with the user's profile and used to provide better search results the next time the user searches.

[1240] Examples and prompts

[1241] As a concrete example, consider a case where a user searches for information on "project management." In this case, the user enters "project management" into the device, and the emotion engine detects that the user is feeling stressed. The device sends the search keywords and emotion data to the server, which then extracts related documents based on the keyword "project management." In this case, a generative AI model is used to generate search results, and documents with titles such as "Project Management Basics" and "Efficient Project Management Methods" are sent to the device. The device displays these results to the user, allowing them to quickly access the information they need.

[1242] Example prompt sentence:

[1243] "In a search system that is integrated with an emotion engine, if a user searches for the keyword 'project management' and detects that they are feeling stressed, what search results should be displayed? Use a generative AI model to suggest the most appropriate search results."

[1244] This system combines a user's search history with emotional data to provide more relevant information that is in line with the user's emotions quickly and accurately, improving the information search experience. Furthermore, continuous learning using machine learning models improves the accuracy of search results from the next time onwards.

[1245] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1246] Step 1:

[1247] Data Indexing

[1248] The server retrieves all document information from the content management system. The input is the document information from the content management system. The server parses and indexes the document information using Apache Lucene. The indexed data is stored in MongoDB, providing an index data store for efficient data searches. When new documents are added, the index is automatically updated.

[1249] Step 2:

[1250] Enter search keywords and sentiment data

[1251] The user enters a search keyword (e.g., "project management") into the search bar and sends it to the device. The input at this time is the search keyword entered by the user. The device uses the built-in camera and microphone to analyze the user's facial expressions and tone of voice to collect emotional data. The emotional data is analyzed using Microsoft Azure's emotion recognition API. This results in the search keyword and the user's emotional data.

[1252] Step 3:

[1253] Submitting a search request

[1254] The device combines the acquired search keywords and emotion data to construct a search request. The input at this time is the search keywords and emotion data. The device uses JavaScript to form the request in JSON format and sends it to the server as an HTTP POST request. This results in a search request to the server.

[1255] Step 4:

[1256] Retrieving relevant documents and using sentiment data

[1257] The server analyzes the received search request and extracts search keywords and sentiment data. The input is the search request, and the output is the search keywords and sentiment data. The server searches the index data store and extracts the most relevant documents based on the search keywords. A generative AI model (e.g., OpenAI's GPT-3) is used for this data extraction. The generative AI model takes sentiment data into account to generate optimal search results, resulting in the most relevant search results.

[1258] Step 5:

[1259] Displaying search results

[1260] The server sends the generated search results to the terminal. The input at this time is the generated search results, and the output is the search result response from the server. The terminal renders the received search results in HTML format and displays them to the user. React.js is used for display. This allows the user to quickly access the desired information by referring to the titles and access paths of related documents.

[1261] Step 6:

[1262] Recording and learning search history and emotion data

[1263] The server records the user's search history and sentiment data in a database (e.g., Apache Hadoop). The input is the user's search history and sentiment data. The output is the search history and sentiment data after it has been saved in the database. This data is used to improve the accuracy of search results from the next time onwards using a machine learning model (e.g., Scikit-learn). This updates the user profile, and more relevant results will be provided for the next search.

[1264] Step 7:

[1265] Using generative AI models

[1266] The server uses a generative AI model to understand the context of the user's search keywords and emotional data. The inputs are the search keywords and emotional data. The output is search results optimized based on the context. The generative AI model (e.g., OpenAI's GPT-3) generates search results that also take emotional data into account. This provides search results that are appropriate for the user's emotions.

[1267] (Application example 2)

[1268] 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."

[1269] Conventional information search systems have difficulty quickly and appropriately providing the necessary information from a large amount of information. Furthermore, they lacked personalization functionality that took into account the user's search trends and emotional state, making it impossible to provide the most useful information for the user. As a result, information search efficiency declined, and user satisfaction was compromised.

[1270] 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 indexing necessary information from a large amount of information, means for receiving search keywords from the user, means for extracting related documents from the indexed data based on the search keywords, means for displaying search results to the user, means for recording the user's search history and emotional data and learning the user's search tendencies and emotional state, and means for personalizing search results based on the user's emotional data. This makes it possible to quickly and accurately provide information related to the user's search keywords and emotional data.

[1271] "A means of indexing necessary information from a large amount of information" is the process by which a server retrieves all document information from a content management system and stores it in an index data store to speed up searches.

[1272] The "means for receiving search keywords from a user" refers to a process in which a user inputs search keywords to search for specific information through a terminal.

[1273] "Means for extracting relevant documents from indexed data based on search keywords" refers to the process by which the server searches the index based on the search keywords and extracts the most relevant documents.

[1274] The "means for displaying search results to the user" refers to the process by which the terminal displays the search results received from the server on the screen, allowing the user to access the desired information.

[1275] "Means for recording users' search history and emotional data and learning their search tendencies and emotional states" refers to the process by which the server records users' past search history and emotional data, and updates the algorithm based on this data to understand users' search tendencies and emotions.

[1276] "Means for personalizing search results based on user emotional data" refers to a process in which the server utilizes the user's emotional data obtained from the emotion engine to optimize search results to adapt to the user's emotional state.

[1277] "Means for updating the machine learning model based on search query input and sentiment data to reflect the user's tendencies and emotional state for future searches" refers to the process by which the server uses the user's search query and sentiment data to train the machine learning model to provide results that adapt to the user's individual tendencies and emotional state for future searches.

[1278] "Means of using a generative AI model to quickly and accurately provide information related to a user's search keywords and sentiment data" refers to the process by which a generative AI model understands a user's search keywords and sentiment data and generates optimal search results.

[1279] The present invention is a system that quickly and accurately searches for necessary information from a large amount of information and provides personalized search results based on the user's search trends and emotions. This system works in cooperation with a server, a terminal, and an emotion engine.

[1280] The server first retrieves all document information from the content management system and indexes them. The indexed data is used to speed up searches and is stored in an index data store. Indexing occurs automatically whenever a new document is added.

[1281] To search for specific information, a user inputs search keywords into the device. The device also has an emotion engine that recognizes the user's emotions, and collects the user's emotion data.

[1282] The terminal receives search keywords and emotion data input by the user and transmits this information to the server.

[1283] The server searches the index based on the received search keywords to extract the most relevant documents, taking into account the user's emotional data and optimizing the search results in a format that is appropriate for the user's emotions.

[1284] The terminal displays the search results received from the server to the user. The displayed search results include titles and paths, which the user can refer to to access the desired information.

[1285] The server records the user's search history and emotion data, associates them with the user profile, and stores them. Furthermore, by updating the machine learning model, the server can provide more relevant search results based on the user's search habits and emotions the next time the user searches.

[1286] The server uses a generative AI model to generate optimal search results based on the user's search keywords and emotional data. Specifically, the generative AI model can generate search results taking into account the user's emotional data.

[1287] For example, consider a case where a user searches for "soothing music." The user enters "soothing music" into the device, and the emotion engine detects "stress." The device sends the search keywords and emotion data to the server. The server extracts documents related to "soothing music" from the indexed data and provides a list of music that helps reduce stress. In this case, the generative AI model uses prompts such as the following to generate optimal search results:

[1288] Example prompt sentence:

[1289] "A user searches for 'soothing music'. The user's emotional state is 'stressed'. Generate a list of music that will help reduce stress."

[1290] This makes it possible to provide information quickly and accurately according to the user's emotional state and search tendencies, thereby realizing a system that contributes to improving user satisfaction.

[1291] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1292] Step 1:

[1293] The server retrieves all document information from the content management system and indexes it. This indexed data is used for faster searches and is stored in an index data store. The input is the document data from the content management system and the output is the indexed data.

[1294] Step 2:

[1295] To search for specific information, a user inputs search keywords into the device. The device also incorporates an emotion engine that recognizes the user's emotions, and simultaneously collects the user's emotional data. The input is the user's search keywords and emotional data, and the output is the integrated information of these data.

[1296] Step 3:

[1297] The terminal receives search keywords and emotion data entered by the user and transmits this information to the server. The input is the search keywords and emotion data, and the output is the data transmission result to the server.

[1298] Step 4:

[1299] The server searches the index data store based on the received search keywords to extract the most relevant documents, taking into account the user's emotional data and optimizing the search results in a format that is appropriate for the user's emotions. The input is the search keywords and emotional data, and the output is personalized search results.

[1300] Step 5:

[1301] The terminal displays the search results received from the server to the user. The displayed search results include a title and a path, which the user can refer to to access the desired information. The input is the search results from the server, and the output is the search results displayed to the user.

[1302] Step 6:

[1303] The server records the user's search history and emotion data, associates them with the user profile, and saves them. Furthermore, by updating the machine learning model, the server can provide more appropriate search results based on the user's search trends and emotions the next time they search. The input is the search history and emotion data, and the output is the updated machine learning model.

[1304] Step 7:

[1305] The server uses a generative AI model to generate optimal search results based on the user's search keywords and emotional data. Specifically, the generative AI model can generate search results taking into account the user's emotional data. The input is the search keywords, emotional data, and the generative AI model's prompt, and the output is personalized search results.

[1306] For example, if a user searches for "soothing music," an example prompt might look like this:

[1307] "A user searches for 'soothing music'. The user's emotional state is 'stressed'. Generate a list of music that will help reduce stress."

[1308] 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.

[1309] 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.

[1310] 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.

[1311] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1312] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1313] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1314] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1315] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1316] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1317] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1318] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1319] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1320] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1321] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1322] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1323] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1324] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1325] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1326] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1327] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1328] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1329] The following is further disclosed regarding the above embodiment.

[1330] (Claim 1)

[1331] A means of indexing necessary information from a large amount of information,

[1332] means for receiving search keywords from a user;

[1333] means for extracting relevant documents from the indexed data based on search keywords;

[1334] a means for displaying search results to a user;

[1335] a means for recording a user's search history and learning search trends;

[1336] A system including:

[1337] (Claim 2)

[1338] 2. The system of claim 1, further comprising means for updating the machine learning model based on the input of a search query to reflect user trends in the next search.

[1339] (Claim 3)

[1340] 10. The system of claim 1, further comprising means for using a generative AI model to quickly and accurately provide information related to a user's search keywords.

[1341] "Example 1"

[1342] (Claim 1)

[1343] A means of indexing necessary information from a large amount of information,

[1344] means for receiving search keywords from a user;

[1345] means for extracting relevant data from the indexed data based on search keywords;

[1346] a means for displaying search results to a user;

[1347] a means for recording a user's search history and learning search trends;

[1348] A means for analyzing the context of entered search keywords using a generative AI model to generate optimal search results;

[1349] A system including:

[1350] (Claim 2)

[1351] 2. The system of claim 1, further comprising means for updating the machine learning model based on the input of a search query to reflect user trends in the next search.

[1352] (Claim 3)

[1353] 10. The system of claim 1, further comprising means for using a generative AI model to quickly and accurately provide information related to a user's search keywords.

[1354] "Application Example 1"

[1355] text

[1356] (Claim 1)

[1357] A means of indexing necessary information from a large amount of information,

[1358] means for receiving search keywords from a user;

[1359] means for extracting relevant documents from the indexed data based on search keywords;

[1360] a means for displaying search results to a user;

[1361] a means for recording a user's search history and learning search trends;

[1362] A content distribution service that recommends personalized content using a user's search history and a machine learning model;

[1363] A system including:

[1364] (Claim 2)

[1365] 2. The system of claim 1, further comprising means for updating the machine learning model based on the input of a search query to reflect user trends in the next search.

[1366] (Claim 3)

[1367] 10. The system of claim 1, further comprising means for using a generative AI model to quickly and accurately provide information related to a user's search keywords.

[1368] "Example 2: Combining Emotion Engines"

[1369] (Claim 1)

[1370] A means of indexing necessary information from a large amount of information,

[1371] means for receiving search keywords from a user;

[1372] A means for analyzing user emotions using the collected emotion data;

[1373] means for extracting relevant documents from the indexed data based on search keywords and sentiment data;

[1374] A means for optimizing search results in a format suited to the user's emotions and displaying them to the user;

[1375] a means for recording a user's search history and sentiment data and learning search trends;

[1376] A means for generating optimal search results using generative AI models; and

[1377] A system including:

[1378] (Claim 2)

[1379] 10. The system of claim 1, further comprising means for updating the machine learning model based on the input of a search query to reflect user trends and sentiment data in subsequent searches.

[1380] (Claim 3)

[1381] 10. The system of claim 1, further comprising means for using a generative AI model to quickly and accurately provide relevant information based on a user's search keywords and sentiment data.

[1382] "Application example 2 when combining emotion engines"

[1383] (Claim 1)

[1384] A means of indexing necessary information from a large amount of information,

[1385] means for receiving search keywords from a user;

[1386] means for extracting relevant documents from the indexed data based on search keywords;

[1387] a means for displaying search results to a user;

[1388] means for recording a user's search history and emotional data to learn the user's search tendencies and emotional states;

[1389] a means for personalizing search results based on user sentiment data;

[1390] A system including:

[1391] (Claim 2)

[1392] 10. The system of claim 1, further comprising means for updating the machine learning model based on the search query input and sentiment data to reflect the user's tendencies and sentiment state in subsequent searches.

[1393] (Claim 3)

[1394] 10. The system of claim 1, further comprising means for using a generative AI model to quickly and accurately provide information related to a user's search keywords and sentiment data. [Explanation of symbols]

[1395] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of indexing necessary information from a large amount of information, means for receiving search keywords from a user; means for extracting relevant documents from the indexed data based on search keywords; a means for displaying search results to a user; a means for recording a user's search history and learning search trends; A system including:

2. The system according to claim 1 , further comprising means for updating the machine learning model based on an input search query to reflect user trends in the next search.

3. 10. The system of claim 1, further comprising means for using a generative AI model to quickly and accurately provide information related to a user's search keywords.

Images