System

Abstract

A system is provided.SOLUTION: The system includes a means for receiving a retrieval keyword inputted by a user, a means for passing the received retrieval keyword to a generation type artificial intelligence to analyze the intention, a means for retrieving a document database based on a related keyword obtained from the generation type artificial intelligence, and a means for providing a retrieval result to the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] ---

[0005] Conventional search systems have the problem that it is difficult to efficiently search for related information due to a single character difference or ambiguous expressions in the keywords entered by the user. As a result, users are unable to quickly access the information they need, which reduces work efficiency. [Means for solving the problem]

[0006] The present invention provides a system that receives search keywords entered by a user, passes the keywords to a generative AI to analyze the user's intent, searches a document database based on related keywords obtained from the generative AI, and provides the user with the search results. This system allows the user to quickly obtain appropriate related information even if some of the keywords entered by the user are incorrect, contributing to improved work efficiency.

[0007] ---

[0008] "Search keywords" are words or phrases that a user enters into a search system.

[0009] "Generative AI" refers to AI that has the ability to analyze input information, understand its meaning and intent, and generate related data.

[0010] "Analyzing intent" is the process of interpreting and understanding the purpose and relevance behind the search terms entered by a user.

[0011] "Related keywords" are words or phrases with broader or more specific relevance that are generated by generative artificial intelligence based on the search keywords entered by the user.

[0012] A "document database" is a collection of documents or data that a search system can access to retrieve related information.

[0013] "Search results" refers to information retrieved from a document database based on the search keywords entered by the user and related keywords generated by generative artificial intelligence.

[0014] The "JSON format" is a lightweight data exchange format for structuring and representing data. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] ---

[0037] This invention relates to a system that uses generative artificial intelligence to search for and efficiently provide relevant information based on search keywords entered by users. The system operates through a process of analyzing the intent of the search keywords and retrieving relevant information from a document database.

[0038] System Overview

[0039] In this system, the user enters search keywords into their device and sends them to a server, which then uses generative artificial intelligence to analyze the intent of the keywords. The server then retrieves related information from a database of materials and provides it to the user. The specific process flow and its operation are explained below.

[0040] Explaining program processing in natural language

[0041] User search keyword input

[0042] The user enters any keyword into the search bar of the terminal. For example, they enter "project progress management template." When the user clicks the search button, the terminal constructs this keyword and prepares to send it to the server.

[0043] Sending keywords to the server

[0044] The device constructs a JSON-formatted request from the entered keywords and sends this request to the server. The JSON-formatted request ensures that the keywords are sent to the server in a properly formatted state.

[0045] Analyze keyword intent

[0046] The server passes the received search keywords to a generative AI, which analyzes the intent of the keywords and infers related contexts and phrases. For example, for a "project progress management template," it generates related keywords such as "project management," "progress status," and "progress report."

[0047] Search the document database

[0048] The server uses the related keywords obtained from the generative artificial intelligence to search the document database, which contains various internal documents, and identifies documents that match or are related to these related keywords. The search can use partial match or full-text search techniques.

[0049] Generating search results

[0050] The server lists relevant documents retrieved from the resource database and generates search results to provide relevant information to the user, including document titles, summaries, and links.

[0051] Sending search results to the server

[0052] The server generates search results and sends them to the device as a JSON response, which includes detailed information about related documents for easy user access.

[0053] Displaying search results

[0054] The terminal analyzes the search results received from the server and displays them on the user interface. The terminal presents the user with a list of related documents in an easy-to-understand format, allowing the user to quickly access the information they need.

[0055] Specific examples

[0056] For example, if the user inputs "template for project progress management," the system performs the following operations.

[0057] 1. The user enters search keywords into the device and clicks the search button.

[0058] 2. The device sends the keyword in JSON format to the server.

[0059] 3. The server passes the keywords to a generative AI for analysis, which generates related keywords such as "project management," "progress status," and "progress report."

[0060] 4. The server searches the resource database to identify relevant documents.

[0061] 5. The server generates a list of search results and sends them to the device in JSON format.

[0062] 6. The device analyzes the search results and displays them in the user interface.

[0063] In this way, the system of the present invention can accurately understand the intent of the keywords entered by the user and quickly provide related information, thereby improving the user's search efficiency and increasing business efficiency.

[0064] The processing flow will be explained below.

[0065] ---

[0066] Step 1:

[0067] The user enters "Project progress management template" in the search bar of the device. The user clicks the search button.

[0068] Step 2:

[0069] The device detects user input and constructs a search request in JSON format that includes the entered keywords. For example, the structure is as follows:

[0070] {

[0071] "search_keyword": "Project progress management template"

[0072] }

[0073] Step 3:

[0074] The device sends the constructed search request to the server, using an HTTP POST request.

[0075] Step 4:

[0076] The server analyzes the search request received from the terminal and obtains the search keywords. For example, the server obtains the keyword "project progress management template."

[0077] Step 5:

[0078] The server calls a generative artificial intelligence (AI) and passes the acquired search keywords to the AI. The AI ​​analyzes the user's intent based on these keywords and generates related keywords. For example, the AI ​​generates the following related keywords:

[0079] [

[0080] "project management",

[0081] "Progress Status",

[0082] "Progress Report"

[0083] ]

[0084] Step 6:

[0085] The server retrieves the generated related keywords and uses them to search a database of documents, which stores business-related document data, for example, using SQL queries or a full-text search engine.

[0086] Step 7:

[0087] The server retrieves search results from the resource database and generates a list of relevant documents. Each document contains information such as a title, summary, and links. For example, the search results look like this:

[0088] {

[0089] "results": [

[0090] {

[0091] "title": "Project Management Template",

[0092] "summary": "Project management template.",

[0093] "link": " / documents / project_management_template.pdf"

[0094] },

[0095] {

[0096] "title": "Progress report example",

[0097] "summary": "Template for reporting project progress.",

[0098] "link": " / documents / progress_report_example.pdf"

[0099] }

[0100] ]

[0101] }

[0102] Step 8:

[0103] The search results generated by the server are sent to the terminal as a JSON format response.

[0104] Step 9:

[0105] The terminal receives the response from the server and analyzes the search results. The analysis results are displayed on the user interface. The user can view the titles, summaries, and links of related documents and access the information they need.

[0106] ---

[0107] Through these processing steps, users can efficiently find the information they need. This system accurately understands the user's intent and quickly provides relevant information, improving the efficiency of the entire search process.

[0108] Example 1

[0109] 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."

[0110] Current search systems have difficulty accurately analyzing the intent of search keywords entered by users and quickly providing relevant information. In particular, if the intent of the keywords is not accurately understood, inappropriate search results are provided, reducing the efficiency with which users can access the information they are looking for. Furthermore, if search results are not properly structured, it takes a long time for users to find the information they need. Furthermore, if the processes for submitting keywords and displaying search results are inconsistent, the performance of the entire system may be degraded.

[0111] 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.

[0112] In this invention, the server includes: means for receiving search keywords entered by a user; means for passing the received search keywords to a generative AI model and analyzing the intent; means for searching a document database based on related keywords obtained from the generative AI model; means for providing search results to the user; means for the terminal to construct the entered keywords as a JSON-formatted request and send it to the server; and means for the server to send the search results to the terminal as a JSON-formatted response. This enables the intent of the keywords entered by the user to be accurately analyzed and related information to be provided efficiently and quickly. Furthermore, exchanging data in a unified format also improves overall system performance.

[0113] "User-entered search keywords" refers to words or phrases that a user enters to search for information.

[0114] A "generative AI model" refers to a machine learning algorithm or artificial intelligence that generates appropriate responses and relevant information from input data.

[0115] "Means of intent analysis" refers to the processes and techniques used to understand and analyze the purpose or intent behind the keywords entered.

[0116] "Related keywords" refer to other keywords or phrases related to the search keywords entered by the user.

[0117] A "document database" refers to a data store that stores various data and documents that can be searched.

[0118] "Search Results" refers to the list of relevant data or documents found through the search process.

[0119] "Means of presentation to users" refers to the processes and techniques used to present search results in a form that users can use.

[0120] "JSON format" is an abbreviation for JavaScript (registered trademark) Object Notation, and refers to a lightweight data exchange format for structuring, storing, and transmitting data.

[0121] A "request" refers to a request message for a service or data sent from a terminal to a server.

[0122] A "response" refers to a reply message sent from a server to a terminal in response to a request.

[0123] "User interface" refers to the screen and interactive environment that allows users to operate a system.

[0124] This invention relates to a system that uses a generative AI model to search for and efficiently provide relevant information based on search keywords entered by users. The system operates through a process of analyzing the intent of the search keywords and retrieving relevant information from a document database.

[0125] System Overview

[0126] The system sends search keywords entered by users to a server, which then uses a generative AI model to analyze the intent of the keywords, retrieving relevant information from a database of materials and providing it to the user.

[0127] Hardware or software used

[0128] The following hardware and software are used to implement this system:

[0129] Device: The device on which a user enters search keywords. Examples include PCs, smartphones, and tablets.

[0130] Server: A central processing unit that analyzes search keywords, searches the document database, and generates results. It can be a cloud server or an on-premise server.

[0131] Generative AI model: An artificial intelligence algorithm that analyzes the intent of keywords and generates related information. For example, a language model such as GPT-3 (registered trademark) is used.

[0132] Material database: A data store that stores the data to be searched. SQL databases, NoSQL databases, ElasticSearch (registered trademark), etc. can be used.

[0133] Explaining program processing in natural language

[0134] User search keyword input

[0135] The user enters any keyword into the search bar of the terminal. For example, they enter "project progress management template." When the user clicks the search button, the terminal constructs this keyword and prepares to send it to the server.

[0136] Sending keywords to the server

[0137] The device constructs a JSON-formatted request from the entered keywords and sends this request to the server. The JSON format ensures that the keywords are sent to the server in a properly formatted state.

[0138] Analyze keyword intent

[0139] The server passes the received search keywords to the generative AI model, which analyzes the intent of the keywords and infers related contexts and phrases. For example, for "project progress management template," it generates related keywords such as "project management," "progress status," and "progress report."

[0140] Search the document database

[0141] The server uses the relevant keywords obtained from the generative AI model to search a database of documents, which contains various documents, to identify documents that match or are related to the relevant keywords. The search can use partial match or full-text search techniques.

[0142] Generating search results

[0143] The server lists relevant documents retrieved from the resource database and generates search results to provide relevant information to the user, including document titles, summaries, and links.

[0144] Sending search results to the server

[0145] The server then sends the generated search results to the device as a JSON response, which includes detailed information about related documents for easy user access.

[0146] Displaying search results

[0147] The terminal analyzes the search results received from the server and displays them on the user interface. The terminal presents the user with a list of related documents in an easy-to-understand format, allowing the user to quickly access the information they need.

[0148] Specific examples

[0149] For example, if the user inputs "template for project progress management," the system performs the following operations.

[0150] 1. The user enters search keywords into the device and clicks the search button.

[0151] 2. The device sends the keyword in JSON format to the server.

[0152] 3. The server passes the keywords to a generative AI model for analysis, generating related keywords such as "project management," "progress status," and "progress report."

[0153] 4. The server searches the resource database to identify relevant documents.

[0154] 5. The server generates a list of search results and sends them to the device in JSON format.

[0155] 6. The device analyzes the search results and displays them in the user interface.

[0156] In this way, the system of the present invention can accurately understand the intent of the keywords entered by the user and quickly provide related information. By analyzing specific phrases such as "project progress management template" as prompts, related information can be accurately obtained. This process improves the user's search efficiency and streamlines work.

[0157] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0158] Step 1:

[0159] The user inputs a search keyword. Specifically, the user inputs any keyword into the search bar of the terminal. As an example of input, the user inputs "template for project progress management." In this way, keywords are collected from the user.

[0160] Step 2:

[0161] The device constructs a request using the keywords in JSON format, for example in JSON format like this:

[0162] json

[0163] {

[0164] "search_query": "project progress management template"

[0165] }

[0166] The terminal constructs this JSON request as an HTTP POST request and sends it to the server. The input is the keyword entered by the user, and the output is the JSON-formatted request.

[0167] Step 3:

[0168] The server parses the received JSON request. The server extracts the value of "search_query" from the request and uses this value in the next step. The input is the JSON formatted request, and the output is the extracted keywords.

[0169] Step 4:

[0170] The server passes the keywords to a generative AI model, which analyzes their intent. The generative AI model is accessed through a specific API. For example, an API request can be used to send a "project progress management template" to the model and retrieve related phrases. The input is the extracted keywords, and the output is a list of related keywords.

[0171] Step 5:

[0172] The server searches the document database based on the related keywords obtained from the generative AI model. For example, it generates a database query using "project management," "progress status," "progress report," etc., and searches the document database. The input is the generated related keywords, and the output is a list of related documents.

[0173] Step 6:

[0174] The server generates a JSON formatted search result list to provide relevant information to the user, for example:

[0175] json

[0176] {

[0177] "results": [

[0178] {

[0179] "title": "Project Management Basics",

[0180] "summary": "Importance of project management and basic techniques",

[0181] "link": "http: / / example.com / doc1"

[0182] },

[0183] {

[0184] "title": "Progress Status Check Template",

[0185] "summary": "Templates for efficiently reviewing and reporting progress",

[0186] "link": "http: / / example.com / doc2"

[0187] }

[0188] ]

[0189] }

[0190] The input is a list of relevant documents, and the output is the search results in JSON format.

[0191] Step 7:

[0192] The server generates search results and sends them to the terminal as a JSON response. The input is the JSON search results and the output is the HTTP response.

[0193] Step 8:

[0194] The terminal analyzes the search results received from the server and displays them on the user interface. Specifically, it analyzes the JSON data of the search results, generates HTML and CSS, and displays them to the user. The input is the JSON-formatted search results, and the output is the information displayed on the user interface.

[0195] Through the above processing steps, a system is realized that accurately understands the intent of the keywords entered by the user and quickly provides related information.

[0196] (Application example 1)

[0197] 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."

[0198] Conventional search systems have the drawback of being inefficient, requiring users to manually search through a vast amount of information to find the data they need. Furthermore, they often fail to accurately understand the intent of search keywords, resulting in irrelevant results being returned. This results in a poor user experience and wastes time and effort.

[0199] 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.

[0200] In this invention, the server includes means for receiving search keywords entered by the user, means for passing the received search keywords to a generative AI and analyzing the intent, means for searching a document database based on related keywords obtained from the generative AI, means for providing search results to the user, and means for controlling the display of search results on the user terminal, thereby enabling the information required by the user to be provided quickly and accurately.

[0201] The "means for receiving search keywords input by a user" is a function that allows the system to receive search keywords input by a user to a terminal.

[0202] "Means for passing received search keywords to a generative AI to analyze the intent" is a function for passing received search keywords to a generative AI and analyzing the intent and purpose of the keywords.

[0203] "Means for searching the document database based on related keywords obtained from generative artificial intelligence" is a function for searching the document database using related keywords extracted from the results of analysis by generative artificial intelligence.

[0204] The "means for providing search results to the user" is a function for presenting search results obtained from the document database to the user.

[0205] The "means for controlling the display of search results on the user terminal" is a function for appropriately displaying search results on the user terminal.

[0206] "Means for sending received search keywords to the server in JSON format" is a function for converting search keywords received from a user into JSON format and sending them to the server.

[0207] "Generative AI infers the user's intent from the search keywords and generates related keywords" is a function in which generative AI infers the user's purpose based on the search keywords and derives highly relevant keywords.

[0208] "Sends related keywords obtained from generative AI to a document database to search for related information, formats the search results, and displays them on a user interface" is a function that searches a document database using related keywords obtained from generative AI, and displays the results in a format that is easy for the user to understand.

[0209] This invention relates to a system that uses generative artificial intelligence to search for and efficiently provide relevant information based on search keywords entered by users. The system operates through a process of analyzing the intent of the search keywords and retrieving relevant information from a document database.

[0210] System Overview

[0211] The system includes the following elements:

[0212] 1. A means of receiving search keywords entered by users

[0213] 2. A method for passing received search keywords to a generative AI to analyze intent

[0214] 3. A method for searching a database of materials based on related keywords obtained from generative artificial intelligence

[0215] 4. How search results are presented to users

[0216] 5. Means of controlling the display of search results on the user's device

[0217] System Operation

[0218] User search keyword input

[0219] The user enters a keyword into the search bar of the device, for example, "benefit list." When the user clicks the search button, the device constructs this keyword and prepares to send it to the server.

[0220] Sending keywords to the server

[0221] The device constructs a JSON-formatted request from the entered keywords and sends this request to the server. The JSON-formatted request ensures that the keywords are sent to the server in a properly formatted state.

[0222] Analyze keyword intent

[0223] The server passes the received search keywords to a generative AI, which analyzes the intent of the keywords and infers related contexts and phrases. For example, for a "benefit list," it generates related keywords such as "benefits," "coupons," and "point redemption."

[0224] Search the document database

[0225] The server uses the related keywords obtained from the generative artificial intelligence to search the resource database, which contains a variety of information, and identifies information that matches or is related to these related keywords. The search can use partial match or full-text search techniques.

[0226] Generating search results

[0227] The server lists related information retrieved from the document database and generates search results for providing the related information to the user, including the title, summary, and link of the information.

[0228] Sending search results to the server

[0229] The server generates search results and sends them to the device as a JSON response, which includes relevant details and makes them easily accessible to the user.

[0230] Displaying search results

[0231] The terminal analyzes the search results received from the server and displays them on the user interface. The terminal presents the user with a list of related information in an easy-to-understand format, allowing the user to quickly access the information they need.

[0232] Specific examples

[0233] For example, if the user types "benefit list," the system will do the following:

[0234] 1. The user enters search keywords into the device and clicks the search button.

[0235] 2. The device sends the keyword in JSON format to the server.

[0236] 3. The server passes the keywords to a generative AI system for analysis, which generates related keywords such as "benefits," "coupons," and "point redemption."

[0237] 4. The server searches the resource database to identify relevant information.

[0238] 5. The server generates a list of search results and sends them to the device in JSON format.

[0239] 6. The device analyzes the search results and displays them in the user interface.

[0240] Example prompts to input to the generative AI model

[0241] Benefits List

[0242] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0243] Step 1:

[0244] The user enters a search keyword into the device's search bar. For example, they enter "benefit list." The entered keyword "benefit list" is temporarily stored in the device's internal memory. When the user clicks the search button, the device begins preparing to send the keyword.

[0245] Step 2:

[0246] The device converts the input keyword "Benefits List" into a JSON request. Specifically, it formats the keyword appropriately and generates the following JSON data:

[0247] json

[0248] {

[0249] "keyword": "Benefits list"

[0250] }

[0251] This JSON formatted request is sent to the server as an HTTP POST request, which allows the server to understand what the user wants to search for.

[0252] Step 3:

[0253] The server analyzes the received JSON data and obtains the search keyword "Benefit List." The server passes this keyword to the generative AI and begins analysis. Specifically, the server inputs the following prompt into the generative AI:

[0254] Benefits List

[0255] Based on this prompt, the generative artificial intelligence analyzes the intent of the keywords and generates related keywords such as "benefits," "coupons," and "point redemption."

[0256] Step 4:

[0257] The server searches the document database based on the related keywords obtained from the generative AI. Specifically, it sends the following search query to the document database:

[0258] json

[0259] {

[0260] "related_keywords": ["benefits", "coupon", "points redemption"]

[0261] }

[0262] The resource database receives this query and searches for relevant information. The server retrieves the list of identified information as search results and formats it.

[0263] Step 5:

[0264] The server generates a list of search results and sends it to the device as a JSON response, which includes a title, summary, and link. An example of a specific response is as follows:

[0265] json

[0266] {

[0267] "results": [

[0268] {"title": "Benefits List", "summary": "Latest Benefits List", "link": "https: / / example.com / benefits"},

[0269] {"title": "Coupon Information", "summary": "Available Coupon Information", "link": "https: / / example.com / coupons"},

[0270] {"title": "Points redemption", "summary": "Detailed information about points redemption", "link": "https: / / example.com / points"}

[0271] ]

[0272] }

[0273] Step 6:

[0274] The device parses the received JSON format search results and displays them on the user interface. Specifically, the parsed search results are displayed as follows:

[0275] 1. Benefits List: Latest Benefits List

[0276] Link: https: / / example.com / benefits

[0277] 2. Coupon Information: Available coupon information

[0278] Link: https: / / example.com / coupons

[0279] 3. Point Redemption: Detailed information on point redemption

[0280] Link: https: / / example.com / points

[0281] The user can view the results and quickly access the information they need.

[0282] Through the above processing steps, related information generated based on the search keyword entered by the user is efficiently presented.

[0283] 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.

[0284] ---

[0285] The present invention relates to a system that uses generative artificial intelligence (AI) to search for and efficiently provide relevant information based on search keywords entered by users. The present invention further improves search accuracy and user experience by combining it with an emotion engine that recognizes user emotions.

[0286] System Overview

[0287] In this system, search keywords entered by the user on their device are sent to a server, which then uses generative artificial intelligence and an emotion engine to analyze the intent of the keywords and the user's emotions. Based on the analysis results, related information is retrieved from a document database and provided to the user. The specific process flow and operation are explained below.

[0288] Explaining program processing in natural language

[0289] User search keyword input

[0290] The user inputs any keyword into the search bar of the terminal. For example, the user inputs "template for project progress management." When the user clicks the search button, the terminal constructs the keyword and prepares to send it to the server.

[0291] Send keywords and emotional information to the server

[0292] The device constructs a search request in JSON format that includes the entered keywords and emotional information obtained from the user's facial expressions and voice, and sends this request to the server. For example, the structure may look like the following:

[0293] {

[0294] "search_keyword": "Project progress management template",

[0295] "emotion_data": {

[0296] "happiness": 0.8,

[0297] "sadness": 0.1,

[0298] "anger": 0.1

[0299] }

[0300] }

[0301] Keyword and sentiment analysis

[0302] The server analyzes the received search keywords and emotional information. The server then passes the search keywords to a generative artificial intelligence (AI) that analyzes their intent. The emotion engine then analyzes the user's emotional information and adjusts related keywords based on that emotion. For example, if the user indicates a high level of happiness, it will prioritize generating positive related keywords.

[0303] Search the document database

[0304] The server uses the related keywords obtained from the generative artificial intelligence and emotion engine to search the document database, which contains various internal company documents, and identifies documents that match or are related to these related keywords. The search can use partial match and full-text search techniques.

[0305] Generating search results

[0306] The server retrieves search results from the resource database and generates a list of relevant documents. Each document contains information such as a title, summary, and links. For example, the search results look like this:

[0307] {

[0308] "results": [

[0309] {

[0310] "title": "Project Management Template",

[0311] "summary": "Project management template.",

[0312] "link": " / documents / project_management_template.pdf"

[0313] },

[0314] {

[0315] "title": "Progress report example",

[0316] "summary": "Template for reporting project progress.",

[0317] "link": " / documents / progress_report_example.pdf"

[0318] }

[0319] ]

[0320] }

[0321] Sending search results to the server

[0322] The server then sends the generated search results to the device as a JSON response. The search results are adjusted to prioritize relevant documents based on emotional information.

[0323] Displaying search results

[0324] The device receives the response from the server and analyzes the search results. The analysis results are displayed in the user interface. The user can view the titles, summaries, and links of related documents and access the information they need. Recommendations based on sentiment may also be displayed.

[0325] Specific examples

[0326] For example, if a user inputs "project progress management template" and simultaneously indicates a high happiness level, the system will perform the following actions:

[0327] 1. The user enters search keywords into the device and clicks the search button.

[0328] 2. The device sends keywords and emotion information in JSON format to the server.

[0329] 3. The server passes the keywords to the generative AI and the emotional information to the emotion engine for analysis.

[0330] 4. Generative AI generates related keywords such as "project management," "progress status," and "progress report."

[0331] 5. The emotion engine prioritizes positive related keywords based on high happiness.

[0332] 6. The server searches the resource database to identify relevant documents.

[0333] 7. The server generates a list of search results and sends them to the device in JSON format.

[0334] 8. The device analyzes the search results and displays them in the user interface.

[0335] In this way, the system of the present invention can accurately grasp the user's intentions and emotions and provide relevant information quickly and appropriately, thereby improving the user's search experience and increasing work efficiency.

[0336] The processing flow will be explained below.

[0337] ---

[0338] Step 1:

[0339] The user enters "Project progress management template" in the search bar of the device. The user clicks the search button.

[0340] Step 2:

[0341] The device detects user input and constructs a search request in JSON format that includes the entered keywords. At the same time, it collects emotional information obtained from the user's facial expressions and voice and adds it to the request. An example structure would be as follows:

[0342] {

[0343] "search_keyword": "Project progress management template",

[0344] "emotion_data": {

[0345] "happiness": 0.8,

[0346] "sadness": 0.1,

[0347] "anger": 0.1

[0348] }

[0349] }

[0350] Step 3:

[0351] The device sends the constructed search request to the server, using an HTTP POST request.

[0352] Step 4:

[0353] The server analyzes the search request received from the device and obtains search keywords and emotion information. For example, the server obtains the keyword "project progress management template" and emotion information.

[0354] Step 5:

[0355] The server calls a generative artificial intelligence (AI) and passes the acquired search keywords to the AI. The AI ​​analyzes the user's intent based on these keywords and generates related keywords. For example, the AI ​​generates the following related keywords:

[0356] [

[0357] "project management",

[0358] "Progress Status",

[0359] "Progress Report"

[0360] ]

[0361] Step 6:

[0362] The server calls the emotion engine and passes the acquired emotion information to the emotion engine. The emotion engine analyzes the user's emotion and adjusts the generated related keywords based on the user's happiness level. For example, if the user indicates a high happiness level, it will prioritize positive related keywords.

[0363] Step 7:

[0364] The server retrieves the adjusted related keywords obtained from the generative artificial intelligence and emotion engine, and uses them to search a document database, which stores business-related document data, using, for example, SQL queries or a full-text search engine.

[0365] Step 8:

[0366] The server retrieves search results from the resource database and generates a list of relevant documents. Each document contains information such as a title, summary, and links. For example, the search results look like this:

[0367] {

[0368] "results": [

[0369] {

[0370] "title": "Project Management Template",

[0371] "summary": "Project management template.",

[0372] "link": " / documents / project_management_template.pdf"

[0373] },

[0374] {

[0375] "title": "Progress report example",

[0376] "summary": "Template for reporting project progress.",

[0377] "link": " / documents / progress_report_example.pdf"

[0378] }

[0379] ]

[0380] }

[0381] Step 9:

[0382] The server generates search results and sends them to the terminal as a JSON response. The search results are adjusted to prioritize relevant documents based on emotional information.

[0383] Step 10:

[0384] The device receives the response from the server and analyzes the search results. The analysis results are displayed in the user interface. The user can view the titles, summaries, and links of related documents and access the information they need. Recommendations based on sentiment may also be displayed.

[0385] ---

[0386] Through these processing steps, users can efficiently find the information they need. This system accurately understands the user's intentions and emotions and provides relevant information quickly and appropriately, improving the efficiency of the entire search process and the user experience.

[0387] Example 2

[0388] 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."

[0389] Conventional information retrieval systems have difficulty accurately grasping the user's search intent, and the search results often do not meet the user's expectations. Furthermore, they provide search results without considering the user's emotions, which results in a poor user experience. It is necessary to solve these problems and provide highly accurate search results that take the user's intent and emotions into account.

[0390] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for acquiring search keywords and emotion information entered by a user in JSON format, means for transmitting the acquired search keywords and emotion information to the server, means for passing the received search keywords to a generative AI model to analyze the intention, means for passing the received emotion information to an emotion analysis engine to evaluate it, means for adjusting related keywords obtained from the generative AI model based on the results of the emotion analysis, means for searching a document database based on the adjusted related keywords, and means for providing search results to the user in JSON format. This makes it possible to provide highly accurate search results that accurately reflect the user's intention and emotion.

[0391] 1. "Search Keywords" are words or phrases that users enter to retrieve information they wish to obtain.

[0392] 2. "Emotional information" is data that represents the user's emotional state obtained from facial expressions, voice, etc.

[0393] 3. "JSON format" is a data format that structures data in text format to make it easier to read.

[0394] 4. "Server" refers to a computer system that receives a request from a client, processes it, and returns a response.

[0395] 5. A "generative AI model" is an artificial intelligence system that generates relevant information or keywords based on specified input.

[0396] 6. "Intent analysis" is the process of understanding the goals and needs behind a user's search keywords.

[0397] 7. "Emotion analysis engine" is a system for analyzing a user's emotional information and evaluating its content.

[0398] 8. “Related Keywords” are additional terms related to the user’s intent that are generated by the generative AI model based on the search keywords.

[0399] 9. "Document Database" means a database containing various information and documents, intended to provide information in response to a search query.

[0400] 10. "Search Results" means a list of information related to the user's search keywords retrieved from the resource database.

[0401] 11. The "HTTP POST method" is the part of the HTTP protocol that allows a client to send data to a server.

[0402] 12. "User interface" means the interface through which a user interacts with a system.

[0403] The present invention relates to a search system that uses a generative AI model and a sentiment analysis engine to search for and efficiently provide relevant information based on search keywords entered by users. The present invention also improves search accuracy and user experience by taking user sentiment into account.

[0404] System Overview

[0405] This system sends search keywords and emotional information entered by users to a server, which then uses a generative AI model and an emotional analysis engine to analyze the intent of the keywords and the user's emotions. Based on the analysis results, it retrieves related information from a document database and provides it to the user.

[0406] Hardware and Software Configuration

[0407] Terminal

[0408] The terminal is a computing device such as a PC, smartphone, or tablet, and provides a means for inputting search keywords and emotion information using a user interface.

[0409] server

[0410] The server is a computer system that receives and processes requests, analyzes search keywords and sentiment information, generates related information, searches the database, and provides results. The following software can be used:

[0411] Generative AI model: Uses a natural language processing model (e.g., GPT-4®) to analyze user intent and generate relevant keywords.

[0412] Sentiment analysis engine: Evaluates the user's emotional information using sentiment analysis technology (e.g., IBM Watson® Emotion Analysis).

[0413] Document database: Document searches are performed using full-text search engines such as Elasticsearch.

[0414] Specific program description

[0415] Input of user search keywords and emotional information

[0416] The user types "project progress management template" into the device's search bar, and the device simultaneously captures emotional information from the user's facial expressions and voice via the device's camera and microphone.

[0417] Sending keywords and sentiment information

[0418] The device constructs the acquired search keywords and emotion information in JSON format and sends it to the server using the HTTP POST method.

[0419] Keyword and emotional analysis

[0420] The server analyzes the received JSON-formatted data and passes the search keywords to the AI ​​model for generation and the emotional information to the emotion analysis engine.

[0421] Generate and adjust related keywords

[0422] The generative AI model generates related keywords based on search keywords. It then adjusts the generated related keywords based on the output of the sentiment analysis engine to create appropriate keywords based on the user's emotional state. For example, if a high level of happiness is indicated, positive keywords will be prioritized.

[0423] Search the document database

[0424] The server uses the tailored related keywords to search a document database, which contains internal and public documents, to identify relevant documents based on the search query.

[0425] Generating and serving search results

[0426] The server creates a JSON-formatted response with the search results and sends it to the device. The device then parses the results and displays them in a user interface. The user can then view the titles, summaries, and links of related documents and access the information they need.

[0427] Specific examples

[0428] For example, if a user types "project progress management template," the following prompt is sent to the generative AI model:

[0429] "Project Progress Management Template"

[0430] If the emotional information is high in happiness, the generated related keywords include "successful project management examples" and "positive progress reports," providing search results that meet user expectations and resulting in a more satisfying search experience.

[0431] In this way, the system of the present invention provides highly accurate search results that reflect the user's intentions and emotions, improving the search experience and increasing work efficiency.

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

[0433] Step 1:

[0434] The user inputs search keywords and emotional information. The user types "project progress management template" into the search bar on their device, and at the same time, emotional information (facial expressions and voice) is acquired via the camera and microphone. The input data obtained is the search keywords in text form and emotional information in numerical form (happiness, sadness, anger, etc.). This becomes the input for the next processing step.

[0435] Step 2:

[0436] The device sends keywords and emotion information to the server. The device then constructs the acquired search keywords and emotion information into a JSON-formatted request and sends it to the server using the HTTP POST method. The input is the search keywords and emotion information, and the output is the JSON-formatted request data.

[0437] Step 3:

[0438] The server analyzes keywords and emotional information. The server first analyzes the received JSON data and separates the search keywords from the emotional information. The server passes the search keywords to the generative AI model and the emotional information to the emotional analysis engine. The input is JSON data, and the output is the provision of individual data to the generative AI model and the emotional analysis engine.

[0439] Step 4:

[0440] The server generates and adjusts related keywords. The generative AI model analyzes search keywords and generates related keywords. Furthermore, the sentiment analysis engine evaluates emotional information and adjusts the generated related keywords. For example, if a high level of happiness is indicated, more positive related keywords are prioritized. The input is data provided to the generative AI model and sentiment analysis engine, and the output is a list of adjusted related keywords.

[0441] Step 5:

[0442] The server searches the document database. Based on the adjusted related keywords, the server queries the document database (e.g., Elasticsearch) to find the relevant documents. The input is the list of adjusted related keywords, and the output is a list of search results.

[0443] Step 6:

[0444] The server generates search results and sends them to the terminal. The server reconstructs the search results obtained from the database into JSON format and sends them to the terminal as an HTTP response. The input is a list of search results, and the output is the response data in JSON format.

[0445] Step 7:

[0446] The terminal displays the search results. The terminal receives the response data from the server, analyzes the search results, and displays them in the user interface. The user can check the titles, summaries, and links of related documents to access the information they need. The input is the response data in JSON format, and the output is the displayed search results.

[0447] In this way, highly accurate search results that reflect the user's intentions and emotions are provided.

[0448] (Application example 2)

[0449] 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."

[0450] Conventional search systems provide search results without considering the user's emotions, which results in poor user experience. Furthermore, they are unable to accurately grasp the user's search intent, making it difficult to provide appropriate and relevant information. Therefore, there is a need for a system that can adjust search results based on the user's emotions and provide more appropriate and relevant information.

[0451] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving search keywords entered by the user, means for transmitting the received search keywords and the user's emotion data to the server in JSON format, means for passing the search keywords and the user's emotion data to a generative AI to analyze the intention and generate related keywords, means for analyzing the user's emotion using an emotion engine and adjusting the related keywords based on the emotion, means for searching a document database, and means for providing the search results to the user. This makes it possible to provide appropriate and highly relevant information that takes the user's emotion into consideration.

[0452] The "means for receiving search keywords input by the user" is an interface for obtaining search keywords input by the user.

[0453] "Means of passing received search keywords to generative artificial intelligence (AI) to analyze intent" refers to the process of sending acquired search keywords to generative artificial intelligence (AI) and analyzing their intent.

[0454] "Means for sending the received search keywords and user emotion data to the server in JSON format" is a function for converting the acquired search keywords and user emotion data into JSON format and sending it to the server.

[0455] "Means for searching a document database based on related keywords obtained from generative artificial intelligence" refers to the process of searching the contents of a document database using related keywords generated by generative artificial intelligence.

[0456] "Means for analyzing user emotions using an emotion engine and adjusting related keywords based on those emotions" refers to a process for analyzing user emotional information using an emotion engine and correcting and adjusting the generated related keywords based on the results.

[0457] "Means for providing search results to users" refers to the interface for organizing the search results and displaying and providing them to users.

[0458] A "document database" is a searchable database in which documents and information are stored.

[0459] This invention is a system that optimizes search results by combining generative artificial intelligence and an emotion engine based on search keywords and emotion data entered by the user.

[0460] System Overview

[0461] Users input search keywords and emotion data using a device (smartphone or head-mounted display). The system sends this data to a server, which then uses generative artificial intelligence and an emotion engine to analyze the user's search intent and emotion. The system then searches for relevant information from a database and provides it to the user.

[0462] Hardware and software used

[0463] Hardware: Smartphone, Head-Mounted Display (HMD)

[0464] Software: Generative AI, emotion engine, JSON processing library (e.g., Python's requests library)

[0465] Operating procedure

[0466] User Input

[0467] The user enters search keywords using the device interface. For example, the user searches for "summer dress." At the same time, the user's emotional data (e.g., happiness, sadness, anger) is also acquired.

[0468] Sending data to the server

[0469] The device converts the search keywords and emotion data into JSON format and sends it to the server as follows:

[0470] {

[0471] "search_keyword": "summer dress",

[0472] "emotion_data": {

[0473] "happiness": 0.9,

[0474] "sadness": 0.0,

[0475] "anger": 0.1

[0476] }

[0477] }

[0478] Intention and emotion analysis

[0479] The server passes the received search keywords and emotion data to a generative AI to analyze the user's intention. The emotion engine also analyzes the user's emotions, and if the emotion is highly positive, it adjusts the related keywords in a more positive direction.

[0480] Search the document database

[0481] Based on the generated related keywords, the server searches the document database for related information such as "summer dress" or "summer dress."

[0482] Providing search results

[0483] The server sends the search results as a JSON response to the terminal, which receives the response and displays it in its user interface.

[0484] Specific examples

[0485] For example, if a user inputs the keyword "summer dress" and the emotional data indicates a high happiness level, the system operates as follows:

[0486] 1. A user enters "summer dress" and the emotion data is "happiness: 0.9, sadness: 0.0, anger: 0.1".

[0487] 2. The device converts this into JSON format and sends it to the server.

[0488] 3. The server passes the received data to the generative AI and emotion engine for analysis.

[0489] 4. Based on the emotional data, the generative AI will prioritize keywords that are related to positive products when you are in a "happy mood."

[0490] 5. The server searches the relevant database of materials to obtain information on positive products.

[0491] 6. The server sends the search results to the device in JSON format.

[0492] 7. The device analyzes the results and displays them in the user interface.

[0493] Prompt Sentence Examples

[0494] "If a user searches for 'summer dress' and the sentiment data looks like this:

[0495] Happiness: 0.9

[0496] Sadness: 0.0

[0497] Anger: 0.1

[0498] In response to this, please propose related products with vibrant and colorful designs.

[0499] This system allows users to get optimal search results tailored to their emotional state, greatly improving the search experience.

[0500] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0501] Step 1:

[0502] The user enters search keywords using the device's interface. For example, the user enters "summer dress." At this time, the device acquires the user's emotional data (happiness, sadness, anger, etc.) along with the search keywords. The inputs are the search keywords in text format and sensor data (emotional data).

[0503] Step 2:

[0504] The device converts the acquired search keywords and emotion data into JSON format and constructs a JSON object like the one below.

[0505] {

[0506] "search_keyword": "summer dress",

[0507] "emotion_data": {

[0508] "happiness": 0.9,

[0509] "sadness": 0.0,

[0510] "anger": 0.1

[0511] }

[0512] }

[0513] This JSON object is sent to the server, which gets the text data and emotion data as input and the constructed JSON object as output.

[0514] Step 3:

[0515] The server parses the received JSON object and extracts search keywords and sentiment data. Specifically, it breaks down the data through JSON parsing. The input is a JSON object, and the output is the extracted text data and sentiment data.

[0516] Step 4:

[0517] The server then passes the extracted search keywords to a generative artificial intelligence (AI) for analysis. This AI then infers the user's intent from the search keywords and executes a process to generate related keywords. The input is text data (search keywords), and the output is a list of related keywords.

[0518] Step 5:

[0519] The server analyzes the user's emotional data using an emotion engine. This emotion engine adjusts the list of related keywords based on the emotional data. The input is the emotional data and the list of related keywords, and the output is the adjusted list of keywords based on the emotions.

[0520] Step 6:

[0521] The server searches the database of materials based on the adjusted keyword list. It uses the adjusted keywords as a search query to retrieve matching or similar materials from the database. The input is the adjusted keyword list, and the output is a list of related materials.

[0522] Step 7:

[0523] The server organizes the acquired data in JSON format and constructs a response like the one below.

[0524] {

[0525] "results": [

[0526] {

[0527] "title": "Summer dress - colorful design",

[0528] "summary": "A colorful and gorgeous summer dress.",

[0529] "link": " / products / summer_dress_1"

[0530] },

[0531] {

[0532] "title": "Summer dress - cool material",

[0533] "summary": "A summer dress made from a refreshing material.",

[0534] "link": " / products / summer_dress_2"

[0535] }

[0536] ]

[0537] }

[0538] The input is a list of documents, and the output is a constructed JSON response.

[0539] Step 8:

[0540] The terminal receives the JSON response from the server, parses it, and displays it in the user interface, including the title, summary, and link of the search result. The input is the JSON response, and the output is the search result displayed on the screen.

[0541] This way, users receive relevant search results based on their emotions.

[0542] 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.

[0543] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0544] 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.

[0545] [Second embodiment]

[0546] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0547] 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.

[0548] 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).

[0549] 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.

[0550] 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.

[0551] 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).

[0552] 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.

[0553] 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.

[0554] 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.

[0555] 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.

[0556] 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.

[0557] 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."

[0558] ---

[0559] This invention relates to a system that uses generative artificial intelligence to search for and efficiently provide relevant information based on search keywords entered by users. The system operates through a process of analyzing the intent of the search keywords and retrieving relevant information from a document database.

[0560] System Overview

[0561] In this system, the user enters search keywords into their device and sends them to a server, which then uses generative artificial intelligence to analyze the intent of the keywords. The server then retrieves related information from a database of materials and provides it to the user. The specific process flow and its operation are explained below.

[0562] Explaining program processing in natural language

[0563] User search keyword input

[0564] The user enters any keyword into the search bar of the terminal. For example, they enter "project progress management template." When the user clicks the search button, the terminal constructs this keyword and prepares to send it to the server.

[0565] Sending keywords to the server

[0566] The device constructs a JSON-formatted request from the entered keywords and sends this request to the server. The JSON-formatted request ensures that the keywords are sent to the server in a properly formatted state.

[0567] Analyze keyword intent

[0568] The server passes the received search keywords to a generative AI, which analyzes the intent of the keywords and infers related contexts and phrases. For example, for a "project progress management template," it generates related keywords such as "project management," "progress status," and "progress report."

[0569] Search the document database

[0570] The server uses the related keywords obtained from the generative artificial intelligence to search the document database, which contains various internal documents, and identifies documents that match or are related to these related keywords. The search can use partial match or full-text search techniques.

[0571] Generating search results

[0572] The server lists relevant documents retrieved from the resource database and generates search results to provide relevant information to the user, including document titles, summaries, and links.

[0573] Sending search results to the server

[0574] The server generates search results and sends them to the device as a JSON response, which includes detailed information about related documents for easy user access.

[0575] Displaying search results

[0576] The terminal analyzes the search results received from the server and displays them on the user interface. The terminal presents the user with a list of related documents in an easy-to-understand format, allowing the user to quickly access the information they need.

[0577] Specific examples

[0578] For example, if the user inputs "template for project progress management," the system performs the following operations.

[0579] 1. The user enters search keywords into the device and clicks the search button.

[0580] 2. The device sends the keyword in JSON format to the server.

[0581] 3. The server passes the keywords to a generative AI for analysis, which generates related keywords such as "project management," "progress status," and "progress report."

[0582] 4. The server searches the resource database to identify relevant documents.

[0583] 5. The server generates a list of search results and sends them to the device in JSON format.

[0584] 6. The device analyzes the search results and displays them in the user interface.

[0585] In this way, the system of the present invention can accurately understand the intent of the keywords entered by the user and quickly provide related information, thereby improving the user's search efficiency and increasing business efficiency.

[0586] The processing flow will be explained below.

[0587] ---

[0588] Step 1:

[0589] The user enters "Project progress management template" in the search bar of the device. The user clicks the search button.

[0590] Step 2:

[0591] The device detects user input and constructs a search request in JSON format that includes the entered keywords. For example, the structure is as follows:

[0592] {

[0593] "search_keyword": "Project progress management template"

[0594] }

[0595] Step 3:

[0596] The device sends the constructed search request to the server, using an HTTP POST request.

[0597] Step 4:

[0598] The server analyzes the search request received from the terminal and obtains the search keywords. For example, the server obtains the keyword "project progress management template."

[0599] Step 5:

[0600] The server calls a generative artificial intelligence (AI) and passes the acquired search keywords to the AI. The AI ​​analyzes the user's intent based on these keywords and generates related keywords. For example, the AI ​​generates the following related keywords:

[0601] [

[0602] "project management",

[0603] "Progress Status",

[0604] "Progress Report"

[0605] ]

[0606] Step 6:

[0607] The server retrieves the generated related keywords and uses them to search a database of documents, which stores business-related document data, for example, using SQL queries or a full-text search engine.

[0608] Step 7:

[0609] The server retrieves search results from the resource database and generates a list of relevant documents. Each document contains information such as a title, summary, and links. For example, the search results look like this:

[0610] {

[0611] "results": [

[0612] {

[0613] "title": "Project Management Template",

[0614] "summary": "Project management template.",

[0615] "link": " / documents / project_management_template.pdf"

[0616] },

[0617] {

[0618] "title": "Progress report example",

[0619] "summary": "Template for reporting project progress.",

[0620] "link": " / documents / progress_report_example.pdf"

[0621] }

[0622] ]

[0623] }

[0624] Step 8:

[0625] The search results generated by the server are sent to the terminal as a JSON format response.

[0626] Step 9:

[0627] The terminal receives the response from the server and analyzes the search results. The analysis results are displayed on the user interface. The user can view the titles, summaries, and links of related documents and access the information they need.

[0628] ---

[0629] Through these processing steps, users can efficiently find the information they need. This system accurately understands the user's intent and quickly provides relevant information, improving the efficiency of the entire search process.

[0630] Example 1

[0631] 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."

[0632] Current search systems have difficulty accurately analyzing the intent of search keywords entered by users and quickly providing relevant information. In particular, if the intent of the keywords is not accurately understood, inappropriate search results are provided, reducing the efficiency with which users can access the information they are looking for. Furthermore, if search results are not properly structured, it takes a long time for users to find the information they need. Furthermore, if the processes for submitting keywords and displaying search results are inconsistent, the performance of the entire system may be degraded.

[0633] 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.

[0634] In this invention, the server includes: means for receiving search keywords entered by a user; means for passing the received search keywords to a generative AI model and analyzing the intent; means for searching a document database based on related keywords obtained from the generative AI model; means for providing search results to the user; means for the terminal to construct the entered keywords as a JSON-formatted request and send it to the server; and means for the server to send the search results to the terminal as a JSON-formatted response. This enables the intent of the keywords entered by the user to be accurately analyzed and related information to be provided efficiently and quickly. Furthermore, exchanging data in a unified format also improves overall system performance.

[0635] "User-entered search keywords" refers to words or phrases that a user enters to search for information.

[0636] A "generative AI model" refers to a machine learning algorithm or artificial intelligence that generates appropriate responses and relevant information from input data.

[0637] "Means of intent analysis" refers to the processes and techniques used to understand and analyze the purpose or intent behind the keywords entered.

[0638] "Related keywords" refer to other keywords or phrases related to the search keywords entered by the user.

[0639] A "document database" refers to a data store that stores various data and documents that can be searched.

[0640] "Search Results" refers to the list of relevant data or documents found through the search process.

[0641] "Means of presentation to users" refers to the processes and techniques used to present search results in a form that users can use.

[0642] "JSON format" is an abbreviation for JavaScript Object Notation, and refers to a lightweight data exchange format for structuring, storing, and transmitting data.

[0643] A "request" refers to a request message for a service or data sent from a terminal to a server.

[0644] A "response" refers to a reply message sent from a server to a terminal in response to a request.

[0645] "User interface" refers to the screen and interactive environment that allows users to operate a system.

[0646] This invention relates to a system that uses a generative AI model to search for and efficiently provide relevant information based on search keywords entered by users. The system operates through a process of analyzing the intent of the search keywords and retrieving relevant information from a document database.

[0647] System Overview

[0648] The system sends search keywords entered by users to a server, which then uses a generative AI model to analyze the intent of the keywords, retrieving relevant information from a database of materials and providing it to the user.

[0649] Hardware or software used

[0650] The following hardware and software are used to implement this system:

[0651] Device: The device on which a user enters search keywords. Examples include PCs, smartphones, and tablets.

[0652] Server: A central processing unit that analyzes search keywords, searches the document database, and generates results. It can be a cloud server or an on-premise server.

[0653] Generative AI model: An artificial intelligence algorithm that analyzes the intent of keywords and generates relevant information. For example, a language model such as GPT-3 is used.

[0654] Material database: A data store that stores the data to be searched. It can be an SQL database, a NoSQL database, ElasticSearch, etc.

[0655] Explaining program processing in natural language

[0656] User search keyword input

[0657] The user enters any keyword into the search bar of the terminal. For example, they enter "project progress management template." When the user clicks the search button, the terminal constructs this keyword and prepares to send it to the server.

[0658] Sending keywords to the server

[0659] The device constructs a JSON-formatted request from the entered keywords and sends this request to the server. The JSON format ensures that the keywords are sent to the server in a properly formatted state.

[0660] Analyze keyword intent

[0661] The server passes the received search keywords to the generative AI model, which analyzes the intent of the keywords and infers related contexts and phrases. For example, for "project progress management template," it generates related keywords such as "project management," "progress status," and "progress report."

[0662] Search the document database

[0663] The server uses the relevant keywords obtained from the generative AI model to search a database of documents, which contains various documents, to identify documents that match or are related to the relevant keywords. The search can use partial match or full-text search techniques.

[0664] Generating search results

[0665] The server lists relevant documents retrieved from the resource database and generates search results to provide relevant information to the user, including document titles, summaries, and links.

[0666] Sending search results to the server

[0667] The server then sends the generated search results to the device as a JSON response, which includes detailed information about related documents for easy user access.

[0668] Displaying search results

[0669] The terminal analyzes the search results received from the server and displays them on the user interface. The terminal presents the user with a list of related documents in an easy-to-understand format, allowing the user to quickly access the information they need.

[0670] Specific examples

[0671] For example, if the user inputs "template for project progress management," the system performs the following operations.

[0672] 1. The user enters search keywords into the device and clicks the search button.

[0673] 2. The device sends the keyword in JSON format to the server.

[0674] 3. The server passes the keywords to a generative AI model for analysis, generating related keywords such as "project management," "progress status," and "progress report."

[0675] 4. The server searches the resource database to identify relevant documents.

[0676] 5. The server generates a list of search results and sends them to the device in JSON format.

[0677] 6. The device analyzes the search results and displays them in the user interface.

[0678] In this way, the system of the present invention can accurately understand the intent of the keywords entered by the user and quickly provide related information. By analyzing specific phrases such as "project progress management template" as prompts, related information can be accurately obtained. This process improves the user's search efficiency and streamlines work.

[0679] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0680] Step 1:

[0681] The user inputs a search keyword. Specifically, the user inputs any keyword into the search bar of the terminal. As an example of input, the user inputs "template for project progress management." In this way, keywords are collected from the user.

[0682] Step 2:

[0683] The device constructs a request using the keywords in JSON format, for example in JSON format like this:

[0684] json

[0685] {

[0686] "search_query": "project progress management template"

[0687] }

[0688] The terminal constructs this JSON request as an HTTP POST request and sends it to the server. The input is the keyword entered by the user, and the output is the JSON-formatted request.

[0689] Step 3:

[0690] The server parses the received JSON request. The server extracts the value of "search_query" from the request and uses this value in the next step. The input is the JSON formatted request, and the output is the extracted keywords.

[0691] Step 4:

[0692] The server passes the keywords to a generative AI model, which analyzes their intent. The generative AI model is accessed through a specific API. For example, an API request can be used to send a "project progress management template" to the model and retrieve related phrases. The input is the extracted keywords, and the output is a list of related keywords.

[0693] Step 5:

[0694] The server searches the document database based on the related keywords obtained from the generative AI model. For example, it generates a database query using "project management," "progress status," "progress report," etc., and searches the document database. The input is the generated related keywords, and the output is a list of related documents.

[0695] Step 6:

[0696] The server generates a JSON formatted search result list to provide relevant information to the user, for example:

[0697] json

[0698] {

[0699] "results": [

[0700] {

[0701] "title": "Project Management Basics",

[0702] "summary": "Importance of project management and basic techniques",

[0703] "link": "http: / / example.com / doc1"

[0704] },

[0705] {

[0706] "title": "Progress Status Check Template",

[0707] "summary": "Templates for efficiently reviewing and reporting progress",

[0708] "link": "http: / / example.com / doc2"

[0709] }

[0710] ]

[0711] }

[0712] The input is a list of relevant documents, and the output is the search results in JSON format.

[0713] Step 7:

[0714] The server generates search results and sends them to the terminal as a JSON response. The input is the JSON search results and the output is the HTTP response.

[0715] Step 8:

[0716] The terminal analyzes the search results received from the server and displays them on the user interface. Specifically, it analyzes the JSON data of the search results, generates HTML and CSS, and displays them to the user. The input is the JSON-formatted search results, and the output is the information displayed on the user interface.

[0717] Through the above processing steps, a system is realized that accurately understands the intent of the keywords entered by the user and quickly provides related information.

[0718] (Application example 1)

[0719] 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."

[0720] Conventional search systems have the drawback of being inefficient, requiring users to manually search through a vast amount of information to find the data they need. Furthermore, they often fail to accurately understand the intent of search keywords, resulting in irrelevant results being returned. This results in a poor user experience and wastes time and effort.

[0721] 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.

[0722] In this invention, the server includes means for receiving search keywords entered by the user, means for passing the received search keywords to a generative AI and analyzing the intent, means for searching a document database based on related keywords obtained from the generative AI, means for providing search results to the user, and means for controlling the display of search results on the user terminal, thereby enabling the information required by the user to be provided quickly and accurately.

[0723] The "means for receiving search keywords input by a user" is a function that allows the system to receive search keywords input by a user to a terminal.

[0724] "Means for passing received search keywords to a generative AI to analyze the intent" is a function for passing received search keywords to a generative AI and analyzing the intent and purpose of the keywords.

[0725] "Means for searching the document database based on related keywords obtained from generative artificial intelligence" is a function for searching the document database using related keywords extracted from the results of analysis by generative artificial intelligence.

[0726] The "means for providing search results to the user" is a function for presenting search results obtained from the document database to the user.

[0727] The "means for controlling the display of search results on the user terminal" is a function for appropriately displaying search results on the user terminal.

[0728] "Means for sending received search keywords to the server in JSON format" is a function for converting search keywords received from a user into JSON format and sending them to the server.

[0729] "Generative AI infers the user's intent from the search keywords and generates related keywords" is a function in which generative AI infers the user's purpose based on the search keywords and derives highly relevant keywords.

[0730] "Sends related keywords obtained from generative AI to a document database to search for related information, formats the search results, and displays them on a user interface" is a function that searches a document database using related keywords obtained from generative AI, and displays the results in a format that is easy for the user to understand.

[0731] This invention relates to a system that uses generative artificial intelligence to search for and efficiently provide relevant information based on search keywords entered by users. The system operates through a process of analyzing the intent of the search keywords and retrieving relevant information from a document database.

[0732] System Overview

[0733] The system includes the following elements:

[0734] 1. A means of receiving search keywords entered by users

[0735] 2. A method for passing received search keywords to a generative AI to analyze intent

[0736] 3. A method for searching a database of materials based on related keywords obtained from generative artificial intelligence

[0737] 4. How search results are presented to users

[0738] 5. Means of controlling the display of search results on the user's device

[0739] System Operation

[0740] User search keyword input

[0741] The user enters a keyword into the search bar of the device, for example, "benefit list." When the user clicks the search button, the device constructs this keyword and prepares to send it to the server.

[0742] Sending keywords to the server

[0743] The device constructs a JSON-formatted request from the entered keywords and sends this request to the server. The JSON-formatted request ensures that the keywords are sent to the server in a properly formatted state.

[0744] Analyze keyword intent

[0745] The server passes the received search keywords to a generative AI, which analyzes the intent of the keywords and infers related contexts and phrases. For example, for a "benefit list," it generates related keywords such as "benefits," "coupons," and "point redemption."

[0746] Search the document database

[0747] The server uses the related keywords obtained from the generative artificial intelligence to search the resource database, which contains a variety of information, and identifies information that matches or is related to these related keywords. The search can use partial match or full-text search techniques.

[0748] Generating search results

[0749] The server lists related information retrieved from the document database and generates search results for providing the related information to the user, including the title, summary, and link of the information.

[0750] Sending search results to the server

[0751] The server generates search results and sends them to the device as a JSON response, which includes relevant details and makes them easily accessible to the user.

[0752] Displaying search results

[0753] The terminal analyzes the search results received from the server and displays them on the user interface. The terminal presents the user with a list of related information in an easy-to-understand format, allowing the user to quickly access the information they need.

[0754] Specific examples

[0755] For example, if the user types "benefit list," the system will do the following:

[0756] 1. The user enters search keywords into the device and clicks the search button.

[0757] 2. The device sends the keyword in JSON format to the server.

[0758] 3. The server passes the keywords to a generative AI system for analysis, which generates related keywords such as "benefits," "coupons," and "point redemption."

[0759] 4. The server searches the resource database to identify relevant information.

[0760] 5. The server generates a list of search results and sends them to the device in JSON format.

[0761] 6. The device analyzes the search results and displays them in the user interface.

[0762] Example prompts to input to the generative AI model

[0763] Benefits List

[0764] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0765] Step 1:

[0766] The user enters a search keyword into the device's search bar. For example, they enter "benefit list." The entered keyword "benefit list" is temporarily stored in the device's internal memory. When the user clicks the search button, the device begins preparing to send the keyword.

[0767] Step 2:

[0768] The device converts the input keyword "Benefits List" into a JSON request. Specifically, it formats the keyword appropriately and generates the following JSON data:

[0769] json

[0770] {

[0771] "keyword": "Benefits list"

[0772] }

[0773] This JSON formatted request is sent to the server as an HTTP POST request, which allows the server to understand what the user wants to search for.

[0774] Step 3:

[0775] The server analyzes the received JSON data and obtains the search keyword "Benefit List." The server passes this keyword to the generative AI and begins analysis. Specifically, the server inputs the following prompt into the generative AI:

[0776] Benefits List

[0777] Based on this prompt, the generative artificial intelligence analyzes the intent of the keywords and generates related keywords such as "benefits," "coupons," and "point redemption."

[0778] Step 4:

[0779] The server searches the document database based on the related keywords obtained from the generative AI. Specifically, it sends the following search query to the document database:

[0780] json

[0781] {

[0782] "related_keywords": ["benefits", "coupon", "points redemption"]

[0783] }

[0784] The resource database receives this query and searches for relevant information. The server retrieves the list of identified information as search results and formats it.

[0785] Step 5:

[0786] The server generates a list of search results and sends it to the device as a JSON response, which includes a title, summary, and link. An example of a specific response is as follows:

[0787] json

[0788] {

[0789] "results": [

[0790] {"title": "Benefits List", "summary": "Latest Benefits List", "link": "https: / / example.com / benefits"},

[0791] {"title": "Coupon Information", "summary": "Available Coupon Information", "link": "https: / / example.com / coupons"},

[0792] {"title": "Points redemption", "summary": "Detailed information about points redemption", "link": "https: / / example.com / points"}

[0793] ]

[0794] }

[0795] Step 6:

[0796] The device parses the received JSON format search results and displays them on the user interface. Specifically, the parsed search results are displayed as follows:

[0797] 1. Benefits List: Latest Benefits List

[0798] Link: https: / / example.com / benefits

[0799] 2. Coupon Information: Available coupon information

[0800] Link: https: / / example.com / coupons

[0801] 3. Point Redemption: Detailed information on point redemption

[0802] Link: https: / / example.com / points

[0803] The user can view the results and quickly access the information they need.

[0804] Through the above processing steps, related information generated based on the search keyword entered by the user is efficiently presented.

[0805] 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.

[0806] ---

[0807] The present invention relates to a system that uses generative artificial intelligence (AI) to search for and efficiently provide relevant information based on search keywords entered by users. The present invention further improves search accuracy and user experience by combining it with an emotion engine that recognizes user emotions.

[0808] System Overview

[0809] In this system, search keywords entered by the user on their device are sent to a server, which then uses generative artificial intelligence and an emotion engine to analyze the intent of the keywords and the user's emotions. Based on the analysis results, related information is retrieved from a document database and provided to the user. The specific process flow and operation are explained below.

[0810] Explaining program processing in natural language

[0811] User search keyword input

[0812] The user inputs any keyword into the search bar of the terminal. For example, the user inputs "template for project progress management." When the user clicks the search button, the terminal constructs the keyword and prepares to send it to the server.

[0813] Send keywords and emotional information to the server

[0814] The device constructs a search request in JSON format that includes the entered keywords and emotional information obtained from the user's facial expressions and voice, and sends this request to the server. For example, the structure may look like the following:

[0815] {

[0816] "search_keyword": "Project progress management template",

[0817] "emotion_data": {

[0818] "happiness": 0.8,

[0819] "sadness": 0.1,

[0820] "anger": 0.1

[0821] }

[0822] }

[0823] Keyword and sentiment analysis

[0824] The server analyzes the received search keywords and emotional information. The server then passes the search keywords to a generative artificial intelligence (AI) that analyzes their intent. The emotion engine then analyzes the user's emotional information and adjusts related keywords based on that emotion. For example, if the user indicates a high level of happiness, it will prioritize generating positive related keywords.

[0825] Search the document database

[0826] The server uses the related keywords obtained from the generative artificial intelligence and emotion engine to search the document database, which contains various internal company documents, and identifies documents that match or are related to these related keywords. The search can use partial match and full-text search techniques.

[0827] Generating search results

[0828] The server retrieves search results from the resource database and generates a list of relevant documents. Each document contains information such as a title, summary, and links. For example, the search results look like this:

[0829] {

[0830] "results": [

[0831] {

[0832] "title": "Project Management Template",

[0833] "summary": "Project management template.",

[0834] "link": " / documents / project_management_template.pdf"

[0835] },

[0836] {

[0837] "title": "Progress report example",

[0838] "summary": "Template for reporting project progress.",

[0839] "link": " / documents / progress_report_example.pdf"

[0840] }

[0841] ]

[0842] }

[0843] Sending search results to the server

[0844] The server then sends the generated search results to the device as a JSON response. The search results are adjusted to prioritize relevant documents based on emotional information.

[0845] Displaying search results

[0846] The device receives the response from the server and analyzes the search results. The analysis results are displayed in the user interface. The user can view the titles, summaries, and links of related documents and access the information they need. Recommendations based on sentiment may also be displayed.

[0847] Specific examples

[0848] For example, if a user inputs "project progress management template" and simultaneously indicates a high happiness level, the system will perform the following actions:

[0849] 1. The user enters search keywords into the device and clicks the search button.

[0850] 2. The device sends keywords and emotion information in JSON format to the server.

[0851] 3. The server passes the keywords to the generative AI and the emotional information to the emotion engine for analysis.

[0852] 4. Generative AI generates related keywords such as "project management," "progress status," and "progress report."

[0853] 5. The emotion engine prioritizes positive related keywords based on high happiness.

[0854] 6. The server searches the resource database to identify relevant documents.

[0855] 7. The server generates a list of search results and sends them to the device in JSON format.

[0856] 8. The device analyzes the search results and displays them in the user interface.

[0857] In this way, the system of the present invention can accurately grasp the user's intentions and emotions and provide relevant information quickly and appropriately, thereby improving the user's search experience and increasing work efficiency.

[0858] The processing flow will be explained below.

[0859] ---

[0860] Step 1:

[0861] The user enters "Project progress management template" in the search bar of the device. The user clicks the search button.

[0862] Step 2:

[0863] The device detects user input and constructs a search request in JSON format that includes the entered keywords. At the same time, it collects emotional information obtained from the user's facial expressions and voice and adds it to the request. An example structure would be as follows:

[0864] {

[0865] "search_keyword": "Project progress management template",

[0866] "emotion_data": {

[0867] "happiness": 0.8,

[0868] "sadness": 0.1,

[0869] "anger": 0.1

[0870] }

[0871] }

[0872] Step 3:

[0873] The device sends the constructed search request to the server, using an HTTP POST request.

[0874] Step 4:

[0875] The server analyzes the search request received from the device and obtains search keywords and emotion information. For example, the server obtains the keyword "project progress management template" and emotion information.

[0876] Step 5:

[0877] The server calls a generative artificial intelligence (AI) and passes the acquired search keywords to the AI. The AI ​​analyzes the user's intent based on these keywords and generates related keywords. For example, the AI ​​generates the following related keywords:

[0878] [

[0879] "project management",

[0880] "Progress Status",

[0881] "Progress Report"

[0882] ]

[0883] Step 6:

[0884] The server calls the emotion engine and passes the acquired emotion information to the emotion engine. The emotion engine analyzes the user's emotion and adjusts the generated related keywords based on the user's happiness level. For example, if the user indicates a high happiness level, it will prioritize positive related keywords.

[0885] Step 7:

[0886] The server retrieves the adjusted related keywords obtained from the generative artificial intelligence and emotion engine, and uses them to search a document database, which stores business-related document data, using, for example, SQL queries or a full-text search engine.

[0887] Step 8:

[0888] The server retrieves search results from the resource database and generates a list of relevant documents. Each document contains information such as a title, summary, and links. For example, the search results look like this:

[0889] {

[0890] "results": [

[0891] {

[0892] "title": "Project Management Template",

[0893] "summary": "Project management template.",

[0894] "link": " / documents / project_management_template.pdf"

[0895] },

[0896] {

[0897] "title": "Progress report example",

[0898] "summary": "Template for reporting project progress.",

[0899] "link": " / documents / progress_report_example.pdf"

[0900] }

[0901] ]

[0902] }

[0903] Step 9:

[0904] The server generates search results and sends them to the terminal as a JSON response. The search results are adjusted to prioritize relevant documents based on emotional information.

[0905] Step 10:

[0906] The device receives the response from the server and analyzes the search results. The analysis results are displayed in the user interface. The user can view the titles, summaries, and links of related documents and access the information they need. Recommendations based on sentiment may also be displayed.

[0907] ---

[0908] Through these processing steps, users can efficiently find the information they need. This system accurately understands the user's intentions and emotions and provides relevant information quickly and appropriately, improving the efficiency of the entire search process and the user experience.

[0909] Example 2

[0910] 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."

[0911] Conventional information retrieval systems have difficulty accurately grasping the user's search intent, and the search results often do not meet the user's expectations. Furthermore, they provide search results without considering the user's emotions, which results in a poor user experience. It is necessary to solve these problems and provide highly accurate search results that take the user's intent and emotions into account.

[0912] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for acquiring search keywords and emotion information entered by a user in JSON format, means for transmitting the acquired search keywords and emotion information to the server, means for passing the received search keywords to a generative AI model to analyze the intention, means for passing the received emotion information to an emotion analysis engine to evaluate it, means for adjusting related keywords obtained from the generative AI model based on the results of the emotion analysis, means for searching a document database based on the adjusted related keywords, and means for providing search results to the user in JSON format. This makes it possible to provide highly accurate search results that accurately reflect the user's intention and emotion.

[0913] 1. "Search Keywords" are words or phrases that users enter to retrieve information they wish to obtain.

[0914] 2. "Emotional information" is data that represents the user's emotional state obtained from facial expressions, voice, etc.

[0915] 3. "JSON format" is a data format that structures data in text format to make it easier to read.

[0916] 4. "Server" refers to a computer system that receives a request from a client, processes it, and returns a response.

[0917] 5. A "generative AI model" is an artificial intelligence system that generates relevant information or keywords based on specified input.

[0918] 6. "Intent analysis" is the process of understanding the goals and needs behind a user's search keywords.

[0919] 7. "Emotion analysis engine" is a system for analyzing a user's emotional information and evaluating its content.

[0920] 8. “Related Keywords” are additional terms related to the user’s intent that are generated by the generative AI model based on the search keywords.

[0921] 9. "Document Database" means a database containing various information and documents, intended to provide information in response to a search query.

[0922] 10. "Search Results" means a list of information related to the user's search keywords retrieved from the resource database.

[0923] 11. The "HTTP POST method" is the part of the HTTP protocol that allows a client to send data to a server.

[0924] 12. "User interface" means the interface through which a user interacts with a system.

[0925] The present invention relates to a search system that uses a generative AI model and a sentiment analysis engine to search for and efficiently provide relevant information based on search keywords entered by users. The present invention also improves search accuracy and user experience by taking user sentiment into account.

[0926] System Overview

[0927] This system sends search keywords and emotional information entered by users to a server, which then uses a generative AI model and an emotional analysis engine to analyze the intent of the keywords and the user's emotions. Based on the analysis results, it retrieves related information from a document database and provides it to the user.

[0928] Hardware and Software Configuration

[0929] Terminal

[0930] The terminal is a computing device such as a PC, smartphone, or tablet, and provides a means for inputting search keywords and emotion information using a user interface.

[0931] server

[0932] The server is a computer system that receives and processes requests, analyzes search keywords and sentiment information, generates related information, searches the database, and provides results. The following software can be used:

[0933] Generative AI models: Use natural language processing models (e.g., GPT-4) to analyze user intent and generate relevant keywords.

[0934] Sentiment analysis engine: Evaluates user emotional information using sentiment analysis technology (e.g., IBM Watson Emotion Analysis).

[0935] Document database: Document searches are performed using full-text search engines such as Elasticsearch.

[0936] Specific program description

[0937] Input of user search keywords and emotional information

[0938] The user types "project progress management template" into the device's search bar, and the device simultaneously captures emotional information from the user's facial expressions and voice via the device's camera and microphone.

[0939] Sending keywords and sentiment information

[0940] The device constructs the acquired search keywords and emotion information in JSON format and sends it to the server using the HTTP POST method.

[0941] Keyword and emotional analysis

[0942] The server analyzes the received JSON-formatted data and passes the search keywords to the AI ​​model for generation and the emotional information to the emotion analysis engine.

[0943] Generate and adjust related keywords

[0944] The generative AI model generates related keywords based on search keywords. It then adjusts the generated related keywords based on the output of the sentiment analysis engine to create appropriate keywords based on the user's emotional state. For example, if a high level of happiness is indicated, positive keywords will be prioritized.

[0945] Search the document database

[0946] The server uses the tailored related keywords to search a document database, which contains internal and public documents, to identify relevant documents based on the search query.

[0947] Generating and serving search results

[0948] The server creates a JSON-formatted response with the search results and sends it to the device. The device then parses the results and displays them in a user interface. The user can then view the titles, summaries, and links of related documents and access the information they need.

[0949] Specific examples

[0950] For example, if a user types "project progress management template," the following prompt is sent to the generative AI model:

[0951] "Project Progress Management Template"

[0952] If the emotional information is high in happiness, the generated related keywords include "successful project management examples" and "positive progress reports," providing search results that meet user expectations and resulting in a more satisfying search experience.

[0953] In this way, the system of the present invention provides highly accurate search results that reflect the user's intentions and emotions, improving the search experience and increasing work efficiency.

[0954] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0955] Step 1:

[0956] The user inputs search keywords and emotional information. The user types "project progress management template" into the search bar on their device, and at the same time, emotional information (facial expressions and voice) is acquired via the camera and microphone. The input data obtained is the search keywords in text form and emotional information in numerical form (happiness, sadness, anger, etc.). This becomes the input for the next processing step.

[0957] Step 2:

[0958] The device sends keywords and emotion information to the server. The device then constructs the acquired search keywords and emotion information into a JSON-formatted request and sends it to the server using the HTTP POST method. The input is the search keywords and emotion information, and the output is the JSON-formatted request data.

[0959] Step 3:

[0960] The server analyzes keywords and emotional information. The server first analyzes the received JSON data and separates the search keywords from the emotional information. The server passes the search keywords to the generative AI model and the emotional information to the emotional analysis engine. The input is JSON data, and the output is the provision of individual data to the generative AI model and the emotional analysis engine.

[0961] Step 4:

[0962] The server generates and adjusts related keywords. The generative AI model analyzes search keywords and generates related keywords. Furthermore, the sentiment analysis engine evaluates emotional information and adjusts the generated related keywords. For example, if a high level of happiness is indicated, more positive related keywords are prioritized. The input is data provided to the generative AI model and sentiment analysis engine, and the output is a list of adjusted related keywords.

[0963] Step 5:

[0964] The server searches the document database. Based on the adjusted related keywords, the server queries the document database (e.g., Elasticsearch) to find the relevant documents. The input is the list of adjusted related keywords, and the output is a list of search results.

[0965] Step 6:

[0966] The server generates search results and sends them to the terminal. The server reconstructs the search results obtained from the database into JSON format and sends them to the terminal as an HTTP response. The input is a list of search results, and the output is the response data in JSON format.

[0967] Step 7:

[0968] The terminal displays the search results. The terminal receives the response data from the server, analyzes the search results, and displays them in the user interface. The user can check the titles, summaries, and links of related documents to access the information they need. The input is the response data in JSON format, and the output is the displayed search results.

[0969] In this way, highly accurate search results that reflect the user's intentions and emotions are provided.

[0970] (Application example 2)

[0971] 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."

[0972] Conventional search systems provide search results without considering the user's emotions, which results in poor user experience. Furthermore, they are unable to accurately grasp the user's search intent, making it difficult to provide appropriate and relevant information. Therefore, there is a need for a system that can adjust search results based on the user's emotions and provide more appropriate and relevant information.

[0973] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving search keywords entered by the user, means for transmitting the received search keywords and the user's emotion data to the server in JSON format, means for passing the search keywords and the user's emotion data to a generative AI to analyze the intention and generate related keywords, means for analyzing the user's emotion using an emotion engine and adjusting the related keywords based on the emotion, means for searching a document database, and means for providing the search results to the user. This makes it possible to provide appropriate and highly relevant information that takes the user's emotion into consideration.

[0974] The "means for receiving search keywords input by the user" is an interface for obtaining search keywords input by the user.

[0975] "Means of passing received search keywords to generative artificial intelligence (AI) to analyze intent" refers to the process of sending acquired search keywords to generative artificial intelligence (AI) and analyzing their intent.

[0976] "Means for sending the received search keywords and user emotion data to the server in JSON format" is a function for converting the acquired search keywords and user emotion data into JSON format and sending it to the server.

[0977] "Means for searching a document database based on related keywords obtained from generative artificial intelligence" refers to the process of searching the contents of a document database using related keywords generated by generative artificial intelligence.

[0978] "Means for analyzing user emotions using an emotion engine and adjusting related keywords based on those emotions" refers to a process for analyzing user emotional information using an emotion engine and correcting and adjusting the generated related keywords based on the results.

[0979] "Means for providing search results to users" refers to the interface for organizing the search results and displaying and providing them to users.

[0980] A "document database" is a searchable database in which documents and information are stored.

[0981] This invention is a system that optimizes search results by combining generative artificial intelligence and an emotion engine based on search keywords and emotion data entered by the user.

[0982] System Overview

[0983] Users input search keywords and emotion data using a device (smartphone or head-mounted display). The system sends this data to a server, which then uses generative artificial intelligence and an emotion engine to analyze the user's search intent and emotion. The system then searches for relevant information from a database and provides it to the user.

[0984] Hardware and software used

[0985] Hardware: Smartphone, Head-Mounted Display (HMD)

[0986] Software: Generative AI, emotion engine, JSON processing library (e.g., Python's requests library)

[0987] Operating procedure

[0988] User Input

[0989] The user enters search keywords using the device interface. For example, the user searches for "summer dress." At the same time, the user's emotional data (e.g., happiness, sadness, anger) is also acquired.

[0990] Sending data to the server

[0991] The device converts the search keywords and emotion data into JSON format and sends it to the server as follows:

[0992] {

[0993] "search_keyword": "summer dress",

[0994] "emotion_data": {

[0995] "happiness": 0.9,

[0996] "sadness": 0.0,

[0997] "anger": 0.1

[0998] }

[0999] }

[1000] Intention and emotion analysis

[1001] The server passes the received search keywords and emotion data to a generative AI to analyze the user's intention. The emotion engine also analyzes the user's emotions, and if the emotion is highly positive, it adjusts the related keywords in a more positive direction.

[1002] Search the document database

[1003] Based on the generated related keywords, the server searches the document database for related information such as "summer dress" or "summer dress."

[1004] Providing search results

[1005] The server sends the search results as a JSON response to the terminal, which receives the response and displays it in its user interface.

[1006] Specific examples

[1007] For example, if a user inputs the keyword "summer dress" and the emotional data indicates a high happiness level, the system operates as follows:

[1008] 1. A user enters "summer dress" and the emotion data is "happiness: 0.9, sadness: 0.0, anger: 0.1".

[1009] 2. The device converts this into JSON format and sends it to the server.

[1010] 3. The server passes the received data to the generative AI and emotion engine for analysis.

[1011] 4. Based on the emotional data, the generative AI will prioritize keywords that are related to positive products when you are in a "happy mood."

[1012] 5. The server searches the relevant database of materials to obtain information on positive products.

[1013] 6. The server sends the search results to the device in JSON format.

[1014] 7. The device analyzes the results and displays them in the user interface.

[1015] Prompt Sentence Examples

[1016] "If a user searches for 'summer dress' and the sentiment data looks like this:

[1017] Happiness: 0.9

[1018] Sadness: 0.0

[1019] Anger: 0.1

[1020] In response to this, please propose related products with vibrant and colorful designs.

[1021] This system allows users to get optimal search results tailored to their emotional state, greatly improving the search experience.

[1022] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1023] Step 1:

[1024] The user enters search keywords using the device's interface. For example, the user enters "summer dress." At this time, the device acquires the user's emotional data (happiness, sadness, anger, etc.) along with the search keywords. The inputs are the search keywords in text format and sensor data (emotional data).

[1025] Step 2:

[1026] The device converts the acquired search keywords and emotion data into JSON format and constructs a JSON object like the one below.

[1027] {

[1028] "search_keyword": "summer dress",

[1029] "emotion_data": {

[1030] "happiness": 0.9,

[1031] "sadness": 0.0,

[1032] "anger": 0.1

[1033] }

[1034] }

[1035] This JSON object is sent to the server, which gets the text data and emotion data as input and the constructed JSON object as output.

[1036] Step 3:

[1037] The server parses the received JSON object and extracts search keywords and sentiment data. Specifically, it breaks down the data through JSON parsing. The input is a JSON object, and the output is the extracted text data and sentiment data.

[1038] Step 4:

[1039] The server then passes the extracted search keywords to a generative artificial intelligence (AI) for analysis. This AI then infers the user's intent from the search keywords and executes a process to generate related keywords. The input is text data (search keywords), and the output is a list of related keywords.

[1040] Step 5:

[1041] The server analyzes the user's emotional data using an emotion engine. This emotion engine adjusts the list of related keywords based on the emotional data. The input is the emotional data and the list of related keywords, and the output is the adjusted list of keywords based on the emotions.

[1042] Step 6:

[1043] The server searches the database of materials based on the adjusted keyword list. It uses the adjusted keywords as a search query to retrieve matching or similar materials from the database. The input is the adjusted keyword list, and the output is a list of related materials.

[1044] Step 7:

[1045] The server organizes the acquired data in JSON format and constructs a response like the one below.

[1046] {

[1047] "results": [

[1048] {

[1049] "title": "Summer dress - colorful design",

[1050] "summary": "A colorful and gorgeous summer dress.",

[1051] "link": " / products / summer_dress_1"

[1052] },

[1053] {

[1054] "title": "Summer dress - cool material",

[1055] "summary": "A summer dress made from a refreshing material.",

[1056] "link": " / products / summer_dress_2"

[1057] }

[1058] ]

[1059] }

[1060] The input is a list of documents, and the output is a constructed JSON response.

[1061] Step 8:

[1062] The terminal receives the JSON response from the server, parses it, and displays it in the user interface, including the title, summary, and link of the search result. The input is the JSON response, and the output is the search result displayed on the screen.

[1063] This way, users receive relevant search results based on their emotions.

[1064] 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.

[1065] 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.

[1066] 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.

[1067] [Third embodiment]

[1068] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1069] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1070] 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).

[1071] 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.

[1072] 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.

[1073] 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).

[1074] 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.

[1075] 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.

[1076] 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.

[1077] 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.

[1078] 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.

[1079] 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."

[1080] ---

[1081] This invention relates to a system that uses generative artificial intelligence to search for and efficiently provide relevant information based on search keywords entered by users. The system operates through a process of analyzing the intent of the search keywords and retrieving relevant information from a document database.

[1082] System Overview

[1083] In this system, the user enters search keywords into their device and sends them to a server, which then uses generative artificial intelligence to analyze the intent of the keywords. The server then retrieves related information from a database of materials and provides it to the user. The specific process flow and its operation are explained below.

[1084] Explaining program processing in natural language

[1085] User search keyword input

[1086] The user enters any keyword into the search bar of the terminal. For example, they enter "project progress management template." When the user clicks the search button, the terminal constructs this keyword and prepares to send it to the server.

[1087] Sending keywords to the server

[1088] The device constructs a JSON-formatted request from the entered keywords and sends this request to the server. The JSON-formatted request ensures that the keywords are sent to the server in a properly formatted state.

[1089] Analyze keyword intent

[1090] The server passes the received search keywords to a generative AI, which analyzes the intent of the keywords and infers related contexts and phrases. For example, for a "project progress management template," it generates related keywords such as "project management," "progress status," and "progress report."

[1091] Search the document database

[1092] The server uses the related keywords obtained from the generative artificial intelligence to search the document database, which contains various internal documents, and identifies documents that match or are related to these related keywords. The search can use partial match or full-text search techniques.

[1093] Generating search results

[1094] The server lists relevant documents retrieved from the resource database and generates search results to provide relevant information to the user, including document titles, summaries, and links.

[1095] Sending search results to the server

[1096] The server generates search results and sends them to the device as a JSON response, which includes detailed information about related documents for easy user access.

[1097] Displaying search results

[1098] The terminal analyzes the search results received from the server and displays them on the user interface. The terminal presents the user with a list of related documents in an easy-to-understand format, allowing the user to quickly access the information they need.

[1099] Specific examples

[1100] For example, if the user inputs "template for project progress management," the system performs the following operations.

[1101] 1. The user enters search keywords into the device and clicks the search button.

[1102] 2. The device sends the keyword in JSON format to the server.

[1103] 3. The server passes the keywords to a generative AI for analysis, which generates related keywords such as "project management," "progress status," and "progress report."

[1104] 4. The server searches the resource database to identify relevant documents.

[1105] 5. The server generates a list of search results and sends them to the device in JSON format.

[1106] 6. The device analyzes the search results and displays them in the user interface.

[1107] In this way, the system of the present invention can accurately understand the intent of the keywords entered by the user and quickly provide related information, thereby improving the user's search efficiency and increasing business efficiency.

[1108] The processing flow will be explained below.

[1109] ---

[1110] Step 1:

[1111] The user enters "Project progress management template" in the search bar of the device. The user clicks the search button.

[1112] Step 2:

[1113] The device detects user input and constructs a search request in JSON format that includes the entered keywords. For example, the structure is as follows:

[1114] {

[1115] "search_keyword": "Project progress management template"

[1116] }

[1117] Step 3:

[1118] The device sends the constructed search request to the server, using an HTTP POST request.

[1119] Step 4:

[1120] The server analyzes the search request received from the terminal and obtains the search keywords. For example, the server obtains the keyword "project progress management template."

[1121] Step 5:

[1122] The server calls a generative artificial intelligence (AI) and passes the acquired search keywords to the AI. The AI ​​analyzes the user's intent based on these keywords and generates related keywords. For example, the AI ​​generates the following related keywords:

[1123] [

[1124] "project management",

[1125] "Progress Status",

[1126] "Progress Report"

[1127] ]

[1128] Step 6:

[1129] The server retrieves the generated related keywords and uses them to search a database of documents, which stores business-related document data, for example, using SQL queries or a full-text search engine.

[1130] Step 7:

[1131] The server retrieves search results from the resource database and generates a list of relevant documents. Each document contains information such as a title, summary, and links. For example, the search results look like this:

[1132] {

[1133] "results": [

[1134] {

[1135] "title": "Project Management Template",

[1136] "summary": "Project management template.",

[1137] "link": " / documents / project_management_template.pdf"

[1138] },

[1139] {

[1140] "title": "Progress report example",

[1141] "summary": "Template for reporting project progress.",

[1142] "link": " / documents / progress_report_example.pdf"

[1143] }

[1144] ]

[1145] }

[1146] Step 8:

[1147] The search results generated by the server are sent to the terminal as a JSON format response.

[1148] Step 9:

[1149] The terminal receives the response from the server and analyzes the search results. The analysis results are displayed on the user interface. The user can view the titles, summaries, and links of related documents and access the information they need.

[1150] ---

[1151] Through these processing steps, users can efficiently find the information they need. This system accurately understands the user's intent and quickly provides relevant information, improving the efficiency of the entire search process.

[1152] Example 1

[1153] 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."

[1154] Current search systems have difficulty accurately analyzing the intent of search keywords entered by users and quickly providing relevant information. In particular, if the intent of the keywords is not accurately understood, inappropriate search results are provided, reducing the efficiency with which users can access the information they are looking for. Furthermore, if search results are not properly structured, it takes a long time for users to find the information they need. Furthermore, if the processes for submitting keywords and displaying search results are inconsistent, the performance of the entire system may be degraded.

[1155] 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.

[1156] In this invention, the server includes: means for receiving search keywords entered by a user; means for passing the received search keywords to a generative AI model and analyzing the intent; means for searching a document database based on related keywords obtained from the generative AI model; means for providing search results to the user; means for the terminal to construct the entered keywords as a JSON-formatted request and send it to the server; and means for the server to send the search results to the terminal as a JSON-formatted response. This enables the intent of the keywords entered by the user to be accurately analyzed and related information to be provided efficiently and quickly. Furthermore, exchanging data in a unified format also improves overall system performance.

[1157] "User-entered search keywords" refers to words or phrases that a user enters to search for information.

[1158] A "generative AI model" refers to a machine learning algorithm or artificial intelligence that generates appropriate responses and relevant information from input data.

[1159] "Means of intent analysis" refers to the processes and techniques used to understand and analyze the purpose or intent behind the keywords entered.

[1160] "Related keywords" refer to other keywords or phrases related to the search keywords entered by the user.

[1161] A "document database" refers to a data store that stores various data and documents that can be searched.

[1162] "Search Results" refers to the list of relevant data or documents found through the search process.

[1163] "Means of presentation to users" refers to the processes and techniques used to present search results in a form that users can use.

[1164] "JSON format" is an abbreviation for JavaScript Object Notation, and refers to a lightweight data exchange format for structuring, storing, and transmitting data.

[1165] A "request" refers to a request message for a service or data sent from a terminal to a server.

[1166] A "response" refers to a reply message sent from a server to a terminal in response to a request.

[1167] "User interface" refers to the screen and interactive environment that allows users to operate a system.

[1168] This invention relates to a system that uses a generative AI model to search for and efficiently provide relevant information based on search keywords entered by users. The system operates through a process of analyzing the intent of the search keywords and retrieving relevant information from a document database.

[1169] System Overview

[1170] The system sends search keywords entered by users to a server, which then uses a generative AI model to analyze the intent of the keywords, retrieving relevant information from a database of materials and providing it to the user.

[1171] Hardware or software used

[1172] The following hardware and software are used to implement this system:

[1173] Device: The device on which a user enters search keywords. Examples include PCs, smartphones, and tablets.

[1174] Server: A central processing unit that analyzes search keywords, searches the document database, and generates results. It can be a cloud server or an on-premise server.

[1175] Generative AI model: An artificial intelligence algorithm that analyzes the intent of keywords and generates relevant information. For example, a language model such as GPT-3 is used.

[1176] Material database: A data store that stores the data to be searched. It can be an SQL database, a NoSQL database, ElasticSearch, etc.

[1177] Explaining program processing in natural language

[1178] User search keyword input

[1179] The user enters any keyword into the search bar of the terminal. For example, they enter "project progress management template." When the user clicks the search button, the terminal constructs this keyword and prepares to send it to the server.

[1180] Sending keywords to the server

[1181] The device constructs a JSON-formatted request from the entered keywords and sends this request to the server. The JSON format ensures that the keywords are sent to the server in a properly formatted state.

[1182] Analyze keyword intent

[1183] The server passes the received search keywords to the generative AI model, which analyzes the intent of the keywords and infers related contexts and phrases. For example, for "project progress management template," it generates related keywords such as "project management," "progress status," and "progress report."

[1184] Search the document database

[1185] The server uses the relevant keywords obtained from the generative AI model to search a database of documents, which contains various documents, to identify documents that match or are related to the relevant keywords. The search can use partial match or full-text search techniques.

[1186] Generating search results

[1187] The server lists relevant documents retrieved from the resource database and generates search results to provide relevant information to the user, including document titles, summaries, and links.

[1188] Sending search results to the server

[1189] The server then sends the generated search results to the device as a JSON response, which includes detailed information about related documents for easy user access.

[1190] Displaying search results

[1191] The terminal analyzes the search results received from the server and displays them on the user interface. The terminal presents the user with a list of related documents in an easy-to-understand format, allowing the user to quickly access the information they need.

[1192] Specific examples

[1193] For example, if the user inputs "template for project progress management," the system performs the following operations.

[1194] 1. The user enters search keywords into the device and clicks the search button.

[1195] 2. The device sends the keyword in JSON format to the server.

[1196] 3. The server passes the keywords to a generative AI model for analysis, generating related keywords such as "project management," "progress status," and "progress report."

[1197] 4. The server searches the resource database to identify relevant documents.

[1198] 5. The server generates a list of search results and sends them to the device in JSON format.

[1199] 6. The device analyzes the search results and displays them in the user interface.

[1200] In this way, the system of the present invention can accurately understand the intent of the keywords entered by the user and quickly provide related information. By analyzing specific phrases such as "project progress management template" as prompts, related information can be accurately obtained. This process improves the user's search efficiency and streamlines work.

[1201] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1202] Step 1:

[1203] The user inputs a search keyword. Specifically, the user inputs any keyword into the search bar of the terminal. As an example of input, the user inputs "template for project progress management." In this way, keywords are collected from the user.

[1204] Step 2:

[1205] The device constructs a request using the keywords in JSON format, for example in JSON format like this:

[1206] json

[1207] {

[1208] "search_query": "project progress management template"

[1209] }

[1210] The terminal constructs this JSON request as an HTTP POST request and sends it to the server. The input is the keyword entered by the user, and the output is the JSON-formatted request.

[1211] Step 3:

[1212] The server parses the received JSON request. The server extracts the value of "search_query" from the request and uses this value in the next step. The input is the JSON formatted request, and the output is the extracted keywords.

[1213] Step 4:

[1214] The server passes the keywords to a generative AI model, which analyzes their intent. The generative AI model is accessed through a specific API. For example, an API request can be used to send a "project progress management template" to the model and retrieve related phrases. The input is the extracted keywords, and the output is a list of related keywords.

[1215] Step 5:

[1216] The server searches the document database based on the related keywords obtained from the generative AI model. For example, it generates a database query using "project management," "progress status," "progress report," etc., and searches the document database. The input is the generated related keywords, and the output is a list of related documents.

[1217] Step 6:

[1218] The server generates a JSON formatted search result list to provide relevant information to the user, for example:

[1219] json

[1220] {

[1221] "results": [

[1222] {

[1223] "title": "Project Management Basics",

[1224] "summary": "Importance of project management and basic techniques",

[1225] "link": "http: / / example.com / doc1"

[1226] },

[1227] {

[1228] "title": "Progress Status Check Template",

[1229] "summary": "Templates for efficiently reviewing and reporting progress",

[1230] "link": "http: / / example.com / doc2"

[1231] }

[1232] ]

[1233] }

[1234] The input is a list of relevant documents, and the output is the search results in JSON format.

[1235] Step 7:

[1236] The server generates search results and sends them to the terminal as a JSON response. The input is the JSON search results and the output is the HTTP response.

[1237] Step 8:

[1238] The terminal analyzes the search results received from the server and displays them on the user interface. Specifically, it analyzes the JSON data of the search results, generates HTML and CSS, and displays them to the user. The input is the JSON-formatted search results, and the output is the information displayed on the user interface.

[1239] Through the above processing steps, a system is realized that accurately understands the intent of the keywords entered by the user and quickly provides related information.

[1240] (Application example 1)

[1241] 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."

[1242] Conventional search systems have the drawback of being inefficient, requiring users to manually search through a vast amount of information to find the data they need. Furthermore, they often fail to accurately understand the intent of search keywords, resulting in irrelevant results being returned. This results in a poor user experience and wastes time and effort.

[1243] 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.

[1244] In this invention, the server includes means for receiving search keywords entered by the user, means for passing the received search keywords to a generative AI and analyzing the intent, means for searching a document database based on related keywords obtained from the generative AI, means for providing search results to the user, and means for controlling the display of search results on the user terminal, thereby enabling the information required by the user to be provided quickly and accurately.

[1245] The "means for receiving search keywords input by a user" is a function that allows the system to receive search keywords input by a user to a terminal.

[1246] "Means for passing received search keywords to a generative AI to analyze the intent" is a function for passing received search keywords to a generative AI and analyzing the intent and purpose of the keywords.

[1247] "Means for searching the document database based on related keywords obtained from generative artificial intelligence" is a function for searching the document database using related keywords extracted from the results of analysis by generative artificial intelligence.

[1248] The "means for providing search results to the user" is a function for presenting search results obtained from the document database to the user.

[1249] The "means for controlling the display of search results on the user terminal" is a function for appropriately displaying search results on the user terminal.

[1250] "Means for sending received search keywords to the server in JSON format" is a function for converting search keywords received from a user into JSON format and sending them to the server.

[1251] "Generative AI infers the user's intent from the search keywords and generates related keywords" is a function in which generative AI infers the user's purpose based on the search keywords and derives highly relevant keywords.

[1252] "Sends related keywords obtained from generative AI to a document database to search for related information, formats the search results, and displays them on a user interface" is a function that searches a document database using related keywords obtained from generative AI, and displays the results in a format that is easy for the user to understand.

[1253] This invention relates to a system that uses generative artificial intelligence to search for and efficiently provide relevant information based on search keywords entered by users. The system operates through a process of analyzing the intent of the search keywords and retrieving relevant information from a document database.

[1254] System Overview

[1255] The system includes the following elements:

[1256] 1. A means of receiving search keywords entered by users

[1257] 2. A method for passing received search keywords to a generative AI to analyze intent

[1258] 3. A method for searching a database of materials based on related keywords obtained from generative artificial intelligence

[1259] 4. How search results are presented to users

[1260] 5. Means of controlling the display of search results on the user's device

[1261] System Operation

[1262] User search keyword input

[1263] The user enters a keyword into the search bar of the device, for example, "benefit list." When the user clicks the search button, the device constructs this keyword and prepares to send it to the server.

[1264] Sending keywords to the server

[1265] The device constructs a JSON-formatted request from the entered keywords and sends this request to the server. The JSON-formatted request ensures that the keywords are sent to the server in a properly formatted state.

[1266] Analyze keyword intent

[1267] The server passes the received search keywords to a generative AI, which analyzes the intent of the keywords and infers related contexts and phrases. For example, for a "benefit list," it generates related keywords such as "benefits," "coupons," and "point redemption."

[1268] Search the document database

[1269] The server uses the related keywords obtained from the generative artificial intelligence to search the resource database, which contains a variety of information, and identifies information that matches or is related to these related keywords. The search can use partial match or full-text search techniques.

[1270] Generating search results

[1271] The server lists related information retrieved from the document database and generates search results for providing the related information to the user, including the title, summary, and link of the information.

[1272] Sending search results to the server

[1273] The server generates search results and sends them to the device as a JSON response, which includes relevant details and makes them easily accessible to the user.

[1274] Displaying search results

[1275] The terminal analyzes the search results received from the server and displays them on the user interface. The terminal presents the user with a list of related information in an easy-to-understand format, allowing the user to quickly access the information they need.

[1276] Specific examples

[1277] For example, if the user types "benefit list," the system will do the following:

[1278] 1. The user enters search keywords into the device and clicks the search button.

[1279] 2. The device sends the keyword in JSON format to the server.

[1280] 3. The server passes the keywords to a generative AI system for analysis, which generates related keywords such as "benefits," "coupons," and "point redemption."

[1281] 4. The server searches the resource database to identify relevant information.

[1282] 5. The server generates a list of search results and sends them to the device in JSON format.

[1283] 6. The device analyzes the search results and displays them in the user interface.

[1284] Example prompts to input to the generative AI model

[1285] Benefits List

[1286] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1287] Step 1:

[1288] The user enters a search keyword into the device's search bar. For example, they enter "benefit list." The entered keyword "benefit list" is temporarily stored in the device's internal memory. When the user clicks the search button, the device begins preparing to send the keyword.

[1289] Step 2:

[1290] The device converts the input keyword "Benefits List" into a JSON request. Specifically, it formats the keyword appropriately and generates the following JSON data:

[1291] json

[1292] {

[1293] "keyword": "Benefits list"

[1294] }

[1295] This JSON formatted request is sent to the server as an HTTP POST request, which allows the server to understand what the user wants to search for.

[1296] Step 3:

[1297] The server analyzes the received JSON data and obtains the search keyword "Benefit List." The server passes this keyword to the generative AI and begins analysis. Specifically, the server inputs the following prompt into the generative AI:

[1298] Benefits List

[1299] Based on this prompt, the generative artificial intelligence analyzes the intent of the keywords and generates related keywords such as "benefits," "coupons," and "point redemption."

[1300] Step 4:

[1301] The server searches the document database based on the related keywords obtained from the generative AI. Specifically, it sends the following search query to the document database:

[1302] json

[1303] {

[1304] "related_keywords": ["benefits", "coupon", "points redemption"]

[1305] }

[1306] The resource database receives this query and searches for relevant information. The server retrieves the list of identified information as search results and formats it.

[1307] Step 5:

[1308] The server generates a list of search results and sends it to the device as a JSON response, which includes a title, summary, and link. An example of a specific response is as follows:

[1309] json

[1310] {

[1311] "results": [

[1312] {"title": "Benefits List", "summary": "Latest Benefits List", "link": "https: / / example.com / benefits"},

[1313] {"title": "Coupon Information", "summary": "Available Coupon Information", "link": "https: / / example.com / coupons"},

[1314] {"title": "Points redemption", "summary": "Detailed information about points redemption", "link": "https: / / example.com / points"}

[1315] ]

[1316] }

[1317] Step 6:

[1318] The device parses the received JSON format search results and displays them on the user interface. Specifically, the parsed search results are displayed as follows:

[1319] 1. Benefits List: Latest Benefits List

[1320] Link: https: / / example.com / benefits

[1321] 2. Coupon Information: Available coupon information

[1322] Link: https: / / example.com / coupons

[1323] 3. Point Redemption: Detailed information on point redemption

[1324] Link: https: / / example.com / points

[1325] The user can view the results and quickly access the information they need.

[1326] Through the above processing steps, related information generated based on the search keyword entered by the user is efficiently presented.

[1327] 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.

[1328] ---

[1329] The present invention relates to a system that uses generative artificial intelligence (AI) to search for and efficiently provide relevant information based on search keywords entered by users. The present invention further improves search accuracy and user experience by combining it with an emotion engine that recognizes user emotions.

[1330] System Overview

[1331] In this system, search keywords entered by the user on their device are sent to a server, which then uses generative artificial intelligence and an emotion engine to analyze the intent of the keywords and the user's emotions. Based on the analysis results, related information is retrieved from a document database and provided to the user. The specific process flow and operation are explained below.

[1332] Explaining program processing in natural language

[1333] User search keyword input

[1334] The user inputs any keyword into the search bar of the terminal. For example, the user inputs "template for project progress management." When the user clicks the search button, the terminal constructs the keyword and prepares to send it to the server.

[1335] Send keywords and emotional information to the server

[1336] The device constructs a search request in JSON format that includes the entered keywords and emotional information obtained from the user's facial expressions and voice, and sends this request to the server. For example, the structure may look like the following:

[1337] {

[1338] "search_keyword": "Project progress management template",

[1339] "emotion_data": {

[1340] "happiness": 0.8,

[1341] "sadness": 0.1,

[1342] "anger": 0.1

[1343] }

[1344] }

[1345] Keyword and sentiment analysis

[1346] The server analyzes the received search keywords and emotional information. The server then passes the search keywords to a generative artificial intelligence (AI) that analyzes their intent. The emotion engine then analyzes the user's emotional information and adjusts related keywords based on that emotion. For example, if the user indicates a high level of happiness, it will prioritize generating positive related keywords.

[1347] Search the document database

[1348] The server uses the related keywords obtained from the generative artificial intelligence and emotion engine to search the document database, which contains various internal company documents, and identifies documents that match or are related to these related keywords. The search can use partial match and full-text search techniques.

[1349] Generating search results

[1350] The server retrieves search results from the resource database and generates a list of relevant documents. Each document contains information such as a title, summary, and links. For example, the search results look like this:

[1351] {

[1352] "results": [

[1353] {

[1354] "title": "Project Management Template",

[1355] "summary": "Project management template.",

[1356] "link": " / documents / project_management_template.pdf"

[1357] },

[1358] {

[1359] "title": "Progress report example",

[1360] "summary": "Template for reporting project progress.",

[1361] "link": " / documents / progress_report_example.pdf"

[1362] }

[1363] ]

[1364] }

[1365] Sending search results to the server

[1366] The server then sends the generated search results to the device as a JSON response. The search results are adjusted to prioritize relevant documents based on emotional information.

[1367] Displaying search results

[1368] The device receives the response from the server and analyzes the search results. The analysis results are displayed in the user interface. The user can view the titles, summaries, and links of related documents and access the information they need. Recommendations based on sentiment may also be displayed.

[1369] Specific examples

[1370] For example, if a user inputs "project progress management template" and simultaneously indicates a high happiness level, the system will perform the following actions:

[1371] 1. The user enters search keywords into the device and clicks the search button.

[1372] 2. The device sends keywords and emotion information in JSON format to the server.

[1373] 3. The server passes the keywords to the generative AI and the emotional information to the emotion engine for analysis.

[1374] 4. Generative AI generates related keywords such as "project management," "progress status," and "progress report."

[1375] 5. The emotion engine prioritizes positive related keywords based on high happiness.

[1376] 6. The server searches the resource database to identify relevant documents.

[1377] 7. The server generates a list of search results and sends them to the device in JSON format.

[1378] 8. The device analyzes the search results and displays them in the user interface.

[1379] In this way, the system of the present invention can accurately grasp the user's intentions and emotions and provide relevant information quickly and appropriately, thereby improving the user's search experience and increasing work efficiency.

[1380] The processing flow will be explained below.

[1381] ---

[1382] Step 1:

[1383] The user enters "Project progress management template" in the search bar of the device. The user clicks the search button.

[1384] Step 2:

[1385] The device detects user input and constructs a search request in JSON format that includes the entered keywords. At the same time, it collects emotional information obtained from the user's facial expressions and voice and adds it to the request. An example structure would be as follows:

[1386] {

[1387] "search_keyword": "Project progress management template",

[1388] "emotion_data": {

[1389] "happiness": 0.8,

[1390] "sadness": 0.1,

[1391] "anger": 0.1

[1392] }

[1393] }

[1394] Step 3:

[1395] The device sends the constructed search request to the server, using an HTTP POST request.

[1396] Step 4:

[1397] The server analyzes the search request received from the device and obtains search keywords and emotion information. For example, the server obtains the keyword "project progress management template" and emotion information.

[1398] Step 5:

[1399] The server calls a generative artificial intelligence (AI) and passes the acquired search keywords to the AI. The AI ​​analyzes the user's intent based on these keywords and generates related keywords. For example, the AI ​​generates the following related keywords:

[1400] [

[1401] "project management",

[1402] "Progress Status",

[1403] "Progress Report"

[1404] ]

[1405] Step 6:

[1406] The server calls the emotion engine and passes the acquired emotion information to the emotion engine. The emotion engine analyzes the user's emotion and adjusts the generated related keywords based on the user's happiness level. For example, if the user indicates a high happiness level, it will prioritize positive related keywords.

[1407] Step 7:

[1408] The server retrieves the adjusted related keywords obtained from the generative artificial intelligence and emotion engine, and uses them to search a document database, which stores business-related document data, using, for example, SQL queries or a full-text search engine.

[1409] Step 8:

[1410] The server retrieves search results from the resource database and generates a list of relevant documents. Each document contains information such as a title, summary, and links. For example, the search results look like this:

[1411] {

[1412] "results": [

[1413] {

[1414] "title": "Project Management Template",

[1415] "summary": "Project management template.",

[1416] "link": " / documents / project_management_template.pdf"

[1417] },

[1418] {

[1419] "title": "Progress report example",

[1420] "summary": "Template for reporting project progress.",

[1421] "link": " / documents / progress_report_example.pdf"

[1422] }

[1423] ]

[1424] }

[1425] Step 9:

[1426] The server generates search results and sends them to the terminal as a JSON response. The search results are adjusted to prioritize relevant documents based on emotional information.

[1427] Step 10:

[1428] The device receives the response from the server and analyzes the search results. The analysis results are displayed in the user interface. The user can view the titles, summaries, and links of related documents and access the information they need. Recommendations based on sentiment may also be displayed.

[1429] ---

[1430] Through these processing steps, users can efficiently find the information they need. This system accurately understands the user's intentions and emotions and provides relevant information quickly and appropriately, improving the efficiency of the entire search process and the user experience.

[1431] Example 2

[1432] 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."

[1433] Conventional information retrieval systems have difficulty accurately grasping the user's search intent, and the search results often do not meet the user's expectations. Furthermore, they provide search results without considering the user's emotions, which results in a poor user experience. It is necessary to solve these problems and provide highly accurate search results that take the user's intent and emotions into account.

[1434] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for acquiring search keywords and emotion information entered by a user in JSON format, means for transmitting the acquired search keywords and emotion information to the server, means for passing the received search keywords to a generative AI model to analyze the intention, means for passing the received emotion information to an emotion analysis engine to evaluate it, means for adjusting related keywords obtained from the generative AI model based on the results of the emotion analysis, means for searching a document database based on the adjusted related keywords, and means for providing search results to the user in JSON format. This makes it possible to provide highly accurate search results that accurately reflect the user's intention and emotion.

[1435] 1. "Search Keywords" are words or phrases that users enter to retrieve information they wish to obtain.

[1436] 2. "Emotional information" is data that represents the user's emotional state obtained from facial expressions, voice, etc.

[1437] 3. "JSON format" is a data format that structures data in text format to make it easier to read.

[1438] 4. "Server" refers to a computer system that receives a request from a client, processes it, and returns a response.

[1439] 5. A "generative AI model" is an artificial intelligence system that generates relevant information or keywords based on specified input.

[1440] 6. "Intent analysis" is the process of understanding the goals and needs behind a user's search keywords.

[1441] 7. "Emotion analysis engine" is a system for analyzing a user's emotional information and evaluating its content.

[1442] 8. “Related Keywords” are additional terms related to the user’s intent that are generated by the generative AI model based on the search keywords.

[1443] 9. "Document Database" means a database containing various information and documents, intended to provide information in response to a search query.

[1444] 10. "Search Results" means a list of information related to the user's search keywords retrieved from the resource database.

[1445] 11. The "HTTP POST method" is the part of the HTTP protocol that allows a client to send data to a server.

[1446] 12. "User interface" means the interface through which a user interacts with a system.

[1447] The present invention relates to a search system that uses a generative AI model and a sentiment analysis engine to search for and efficiently provide relevant information based on search keywords entered by users. The present invention also improves search accuracy and user experience by taking user sentiment into account.

[1448] System Overview

[1449] This system sends search keywords and emotional information entered by users to a server, which then uses a generative AI model and an emotional analysis engine to analyze the intent of the keywords and the user's emotions. Based on the analysis results, it retrieves related information from a document database and provides it to the user.

[1450] Hardware and Software Configuration

[1451] Terminal

[1452] The terminal is a computing device such as a PC, smartphone, or tablet, and provides a means for inputting search keywords and emotion information using a user interface.

[1453] server

[1454] The server is a computer system that receives and processes requests, analyzes search keywords and sentiment information, generates related information, searches the database, and provides results. The following software can be used:

[1455] Generative AI models: Use natural language processing models (e.g., GPT-4) to analyze user intent and generate relevant keywords.

[1456] Sentiment analysis engine: Evaluates user emotional information using sentiment analysis technology (e.g., IBM Watson Emotion Analysis).

[1457] Document database: Document searches are performed using full-text search engines such as Elasticsearch.

[1458] Specific program description

[1459] Input of user search keywords and emotional information

[1460] The user types "project progress management template" into the device's search bar, and the device simultaneously captures emotional information from the user's facial expressions and voice via the device's camera and microphone.

[1461] Sending keywords and sentiment information

[1462] The device constructs the acquired search keywords and emotion information in JSON format and sends it to the server using the HTTP POST method.

[1463] Keyword and emotional analysis

[1464] The server analyzes the received JSON-formatted data and passes the search keywords to the AI ​​model for generation and the emotional information to the emotion analysis engine.

[1465] Generate and adjust related keywords

[1466] The generative AI model generates related keywords based on search keywords. It then adjusts the generated related keywords based on the output of the sentiment analysis engine to create appropriate keywords based on the user's emotional state. For example, if a high level of happiness is indicated, positive keywords will be prioritized.

[1467] Search the document database

[1468] The server uses the tailored related keywords to search a document database, which contains internal and public documents, to identify relevant documents based on the search query.

[1469] Generating and serving search results

[1470] The server creates a JSON-formatted response with the search results and sends it to the device. The device then parses the results and displays them in a user interface. The user can then view the titles, summaries, and links of related documents and access the information they need.

[1471] Specific examples

[1472] For example, if a user types "project progress management template," the following prompt is sent to the generative AI model:

[1473] "Project Progress Management Template"

[1474] If the emotional information is high in happiness, the generated related keywords include "successful project management examples" and "positive progress reports," providing search results that meet user expectations and resulting in a more satisfying search experience.

[1475] In this way, the system of the present invention provides highly accurate search results that reflect the user's intentions and emotions, improving the search experience and increasing work efficiency.

[1476] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1477] Step 1:

[1478] The user inputs search keywords and emotional information. The user types "project progress management template" into the search bar on their device, and at the same time, emotional information (facial expressions and voice) is acquired via the camera and microphone. The input data obtained is the search keywords in text form and emotional information in numerical form (happiness, sadness, anger, etc.). This becomes the input for the next processing step.

[1479] Step 2:

[1480] The device sends keywords and emotion information to the server. The device then constructs the acquired search keywords and emotion information into a JSON-formatted request and sends it to the server using the HTTP POST method. The input is the search keywords and emotion information, and the output is the JSON-formatted request data.

[1481] Step 3:

[1482] The server analyzes keywords and emotional information. The server first analyzes the received JSON data and separates the search keywords from the emotional information. The server passes the search keywords to the generative AI model and the emotional information to the emotional analysis engine. The input is JSON data, and the output is the provision of individual data to the generative AI model and the emotional analysis engine.

[1483] Step 4:

[1484] The server generates and adjusts related keywords. The generative AI model analyzes search keywords and generates related keywords. Furthermore, the sentiment analysis engine evaluates emotional information and adjusts the generated related keywords. For example, if a high level of happiness is indicated, more positive related keywords are prioritized. The input is data provided to the generative AI model and sentiment analysis engine, and the output is a list of adjusted related keywords.

[1485] Step 5:

[1486] The server searches the document database. Based on the adjusted related keywords, the server queries the document database (e.g., Elasticsearch) to find the relevant documents. The input is the list of adjusted related keywords, and the output is a list of search results.

[1487] Step 6:

[1488] The server generates search results and sends them to the terminal. The server reconstructs the search results obtained from the database into JSON format and sends them to the terminal as an HTTP response. The input is a list of search results, and the output is the response data in JSON format.

[1489] Step 7:

[1490] The terminal displays the search results. The terminal receives the response data from the server, analyzes the search results, and displays them in the user interface. The user can check the titles, summaries, and links of related documents to access the information they need. The input is the response data in JSON format, and the output is the displayed search results.

[1491] In this way, highly accurate search results that reflect the user's intentions and emotions are provided.

[1492] (Application example 2)

[1493] 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."

[1494] Conventional search systems provide search results without considering the user's emotions, which results in poor user experience. Furthermore, they are unable to accurately grasp the user's search intent, making it difficult to provide appropriate and relevant information. Therefore, there is a need for a system that can adjust search results based on the user's emotions and provide more appropriate and relevant information.

[1495] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving search keywords entered by the user, means for transmitting the received search keywords and the user's emotion data to the server in JSON format, means for passing the search keywords and the user's emotion data to a generative AI to analyze the intention and generate related keywords, means for analyzing the user's emotion using an emotion engine and adjusting the related keywords based on the emotion, means for searching a document database, and means for providing the search results to the user. This makes it possible to provide appropriate and highly relevant information that takes the user's emotion into consideration.

[1496] The "means for receiving search keywords input by the user" is an interface for obtaining search keywords input by the user.

[1497] "Means of passing received search keywords to generative artificial intelligence (AI) to analyze intent" refers to the process of sending acquired search keywords to generative artificial intelligence (AI) and analyzing their intent.

[1498] "Means for sending the received search keywords and user emotion data to the server in JSON format" is a function for converting the acquired search keywords and user emotion data into JSON format and sending it to the server.

[1499] "Means for searching a document database based on related keywords obtained from generative artificial intelligence" refers to the process of searching the contents of a document database using related keywords generated by generative artificial intelligence.

[1500] "Means for analyzing user emotions using an emotion engine and adjusting related keywords based on those emotions" refers to a process for analyzing user emotional information using an emotion engine and correcting and adjusting the generated related keywords based on the results.

[1501] "Means for providing search results to users" refers to the interface for organizing the search results and displaying and providing them to users.

[1502] A "document database" is a searchable database in which documents and information are stored.

[1503] This invention is a system that optimizes search results by combining generative artificial intelligence and an emotion engine based on search keywords and emotion data entered by the user.

[1504] System Overview

[1505] Users input search keywords and emotion data using a device (smartphone or head-mounted display). The system sends this data to a server, which then uses generative artificial intelligence and an emotion engine to analyze the user's search intent and emotion. The system then searches for relevant information from a database and provides it to the user.

[1506] Hardware and software used

[1507] Hardware: Smartphone, Head-Mounted Display (HMD)

[1508] Software: Generative AI, emotion engine, JSON processing library (e.g., Python's requests library)

[1509] Operating procedure

[1510] User Input

[1511] The user enters search keywords using the device interface. For example, the user searches for "summer dress." At the same time, the user's emotional data (e.g., happiness, sadness, anger) is also acquired.

[1512] Sending data to the server

[1513] The device converts the search keywords and emotion data into JSON format and sends it to the server as follows:

[1514] {

[1515] "search_keyword": "summer dress",

[1516] "emotion_data": {

[1517] "happiness": 0.9,

[1518] "sadness": 0.0,

[1519] "anger": 0.1

[1520] }

[1521] }

[1522] Intention and emotion analysis

[1523] The server passes the received search keywords and emotion data to a generative AI to analyze the user's intention. The emotion engine also analyzes the user's emotions, and if the emotion is highly positive, it adjusts the related keywords in a more positive direction.

[1524] Search the document database

[1525] Based on the generated related keywords, the server searches the document database for related information such as "summer dress" or "summer dress."

[1526] Providing search results

[1527] The server sends the search results as a JSON response to the terminal, which receives the response and displays it in its user interface.

[1528] Specific examples

[1529] For example, if a user inputs the keyword "summer dress" and the emotional data indicates a high happiness level, the system operates as follows:

[1530] 1. A user enters "summer dress" and the emotion data is "happiness: 0.9, sadness: 0.0, anger: 0.1".

[1531] 2. The device converts this into JSON format and sends it to the server.

[1532] 3. The server passes the received data to the generative AI and emotion engine for analysis.

[1533] 4. Based on the emotional data, the generative AI will prioritize keywords that are related to positive products when you are in a "happy mood."

[1534] 5. The server searches the relevant database of materials to obtain information on positive products.

[1535] 6. The server sends the search results to the device in JSON format.

[1536] 7. The device analyzes the results and displays them in the user interface.

[1537] Prompt Sentence Examples

[1538] "If a user searches for 'summer dress' and the sentiment data looks like this:

[1539] Happiness: 0.9

[1540] Sadness: 0.0

[1541] Anger: 0.1

[1542] In response to this, please propose related products with vibrant and colorful designs.

[1543] This system allows users to get optimal search results tailored to their emotional state, greatly improving the search experience.

[1544] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1545] Step 1:

[1546] The user enters search keywords using the device's interface. For example, the user enters "summer dress." At this time, the device acquires the user's emotional data (happiness, sadness, anger, etc.) along with the search keywords. The inputs are the search keywords in text format and sensor data (emotional data).

[1547] Step 2:

[1548] The device converts the acquired search keywords and emotion data into JSON format and constructs a JSON object like the one below.

[1549] {

[1550] "search_keyword": "summer dress",

[1551] "emotion_data": {

[1552] "happiness": 0.9,

[1553] "sadness": 0.0,

[1554] "anger": 0.1

[1555] }

[1556] }

[1557] This JSON object is sent to the server, which gets the text data and emotion data as input and the constructed JSON object as output.

[1558] Step 3:

[1559] The server parses the received JSON object and extracts search keywords and sentiment data. Specifically, it breaks down the data through JSON parsing. The input is a JSON object, and the output is the extracted text data and sentiment data.

[1560] Step 4:

[1561] The server then passes the extracted search keywords to a generative artificial intelligence (AI) for analysis. This AI then infers the user's intent from the search keywords and executes a process to generate related keywords. The input is text data (search keywords), and the output is a list of related keywords.

[1562] Step 5:

[1563] The server analyzes the user's emotional data using an emotion engine. This emotion engine adjusts the list of related keywords based on the emotional data. The input is the emotional data and the list of related keywords, and the output is the adjusted list of keywords based on the emotions.

[1564] Step 6:

[1565] The server searches the database of materials based on the adjusted keyword list. It uses the adjusted keywords as a search query to retrieve matching or similar materials from the database. The input is the adjusted keyword list, and the output is a list of related materials.

[1566] Step 7:

[1567] The server organizes the acquired data in JSON format and constructs a response like the one below.

[1568] {

[1569] "results": [

[1570] {

[1571] "title": "Summer dress - colorful design",

[1572] "summary": "A colorful and gorgeous summer dress.",

[1573] "link": " / products / summer_dress_1"

[1574] },

[1575] {

[1576] "title": "Summer dress - cool material",

[1577] "summary": "A summer dress made from a refreshing material.",

[1578] "link": " / products / summer_dress_2"

[1579] }

[1580] ]

[1581] }

[1582] The input is a list of documents, and the output is a constructed JSON response.

[1583] Step 8:

[1584] The terminal receives the JSON response from the server, parses it, and displays it in the user interface, including the title, summary, and link of the search result. The input is the JSON response, and the output is the search result displayed on the screen.

[1585] This way, users receive relevant search results based on their emotions.

[1586] 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.

[1587] 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.

[1588] 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.

[1589] [Fourth embodiment]

[1590] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1591] 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.

[1592] 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).

[1593] 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.

[1594] 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.

[1595] 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).

[1596] 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.

[1597] 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.

[1598] 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.

[1599] 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.

[1600] 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.

[1601] 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.

[1602] 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."

[1603] ---

[1604] This invention relates to a system that uses generative artificial intelligence to search for and efficiently provide relevant information based on search keywords entered by users. The system operates through a process of analyzing the intent of the search keywords and retrieving relevant information from a document database.

[1605] System Overview

[1606] In this system, the user enters search keywords into their device and sends them to a server, which then uses generative artificial intelligence to analyze the intent of the keywords. The server then retrieves related information from a database of materials and provides it to the user. The specific process flow and its operation are explained below.

[1607] Explaining program processing in natural language

[1608] User search keyword input

[1609] The user enters any keyword into the search bar of the terminal. For example, they enter "project progress management template." When the user clicks the search button, the terminal constructs this keyword and prepares to send it to the server.

[1610] Sending keywords to the server

[1611] The device constructs a JSON-formatted request from the entered keywords and sends this request to the server. The JSON-formatted request ensures that the keywords are sent to the server in a properly formatted state.

[1612] Analyze keyword intent

[1613] The server passes the received search keywords to a generative AI, which analyzes the intent of the keywords and infers related contexts and phrases. For example, for a "project progress management template," it generates related keywords such as "project management," "progress status," and "progress report."

[1614] Search the document database

[1615] The server uses the related keywords obtained from the generative artificial intelligence to search the document database, which contains various internal documents, and identifies documents that match or are related to these related keywords. The search can use partial match or full-text search techniques.

[1616] Generating search results

[1617] The server lists relevant documents retrieved from the resource database and generates search results to provide relevant information to the user, including document titles, summaries, and links.

[1618] Sending search results to the server

[1619] The server generates search results and sends them to the device as a JSON response, which includes detailed information about related documents for easy user access.

[1620] Displaying search results

[1621] The terminal analyzes the search results received from the server and displays them on the user interface. The terminal presents the user with a list of related documents in an easy-to-understand format, allowing the user to quickly access the information they need.

[1622] Specific examples

[1623] For example, if the user inputs "template for project progress management," the system performs the following operations.

[1624] 1. The user enters search keywords into the device and clicks the search button.

[1625] 2. The device sends the keyword in JSON format to the server.

[1626] 3. The server passes the keywords to a generative AI for analysis, which generates related keywords such as "project management," "progress status," and "progress report."

[1627] 4. The server searches the resource database to identify relevant documents.

[1628] 5. The server generates a list of search results and sends them to the device in JSON format.

[1629] 6. The device analyzes the search results and displays them in the user interface.

[1630] In this way, the system of the present invention can accurately understand the intent of the keywords entered by the user and quickly provide related information, thereby improving the user's search efficiency and increasing business efficiency.

[1631] The processing flow will be explained below.

[1632] ---

[1633] Step 1:

[1634] The user enters "Project progress management template" in the search bar of the device. The user clicks the search button.

[1635] Step 2:

[1636] The device detects user input and constructs a search request in JSON format that includes the entered keywords. For example, the structure is as follows:

[1637] {

[1638] "search_keyword": "Project progress management template"

[1639] }

[1640] Step 3:

[1641] The device sends the constructed search request to the server, using an HTTP POST request.

[1642] Step 4:

[1643] The server analyzes the search request received from the terminal and obtains the search keywords. For example, the server obtains the keyword "project progress management template."

[1644] Step 5:

[1645] The server calls a generative artificial intelligence (AI) and passes the acquired search keywords to the AI. The AI ​​analyzes the user's intent based on these keywords and generates related keywords. For example, the AI ​​generates the following related keywords:

[1646] [

[1647] "project management",

[1648] "Progress Status",

[1649] "Progress Report"

[1650] ]

[1651] Step 6:

[1652] The server retrieves the generated related keywords and uses them to search a database of documents, which stores business-related document data, for example, using SQL queries or a full-text search engine.

[1653] Step 7:

[1654] The server retrieves search results from the resource database and generates a list of relevant documents. Each document contains information such as a title, summary, and links. For example, the search results look like this:

[1655] {

[1656] "results": [

[1657] {

[1658] "title": "Project Management Template",

[1659] "summary": "Project management template.",

[1660] "link": " / documents / project_management_template.pdf"

[1661] },

[1662] {

[1663] "title": "Progress report example",

[1664] "summary": "Template for reporting project progress.",

[1665] "link": " / documents / progress_report_example.pdf"

[1666] }

[1667] ]

[1668] }

[1669] Step 8:

[1670] The search results generated by the server are sent to the terminal as a JSON format response.

[1671] Step 9:

[1672] The terminal receives the response from the server and analyzes the search results. The analysis results are displayed on the user interface. The user can view the titles, summaries, and links of related documents and access the information they need.

[1673] ---

[1674] Through these processing steps, users can efficiently find the information they need. This system accurately understands the user's intent and quickly provides relevant information, improving the efficiency of the entire search process.

[1675] Example 1

[1676] 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."

[1677] Current search systems have difficulty accurately analyzing the intent of search keywords entered by users and quickly providing relevant information. In particular, if the intent of the keywords is not accurately understood, inappropriate search results are provided, reducing the efficiency with which users can access the information they are looking for. Furthermore, if search results are not properly structured, it takes a long time for users to find the information they need. Furthermore, if the processes for submitting keywords and displaying search results are inconsistent, the performance of the entire system may be degraded.

[1678] 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.

[1679] In this invention, the server includes: means for receiving search keywords entered by a user; means for passing the received search keywords to a generative AI model and analyzing the intent; means for searching a document database based on related keywords obtained from the generative AI model; means for providing search results to the user; means for the terminal to construct the entered keywords as a JSON-formatted request and send it to the server; and means for the server to send the search results to the terminal as a JSON-formatted response. This enables the intent of the keywords entered by the user to be accurately analyzed and related information to be provided efficiently and quickly. Furthermore, exchanging data in a unified format also improves overall system performance.

[1680] "User-entered search keywords" refers to words or phrases that a user enters to search for information.

[1681] A "generative AI model" refers to a machine learning algorithm or artificial intelligence that generates appropriate responses and relevant information from input data.

[1682] "Means of intent analysis" refers to the processes and techniques used to understand and analyze the purpose or intent behind the keywords entered.

[1683] "Related keywords" refer to other keywords or phrases related to the search keywords entered by the user.

[1684] A "document database" refers to a data store that stores various data and documents that can be searched.

[1685] "Search Results" refers to the list of relevant data or documents found through the search process.

[1686] "Means of presentation to users" refers to the processes and techniques used to present search results in a form that users can use.

[1687] "JSON format" is an abbreviation for JavaScript Object Notation, and refers to a lightweight data exchange format for structuring, storing, and transmitting data.

[1688] A "request" refers to a request message for a service or data sent from a terminal to a server.

[1689] A "response" refers to a reply message sent from a server to a terminal in response to a request.

[1690] "User interface" refers to the screen and interactive environment that allows users to operate a system.

[1691] This invention relates to a system that uses a generative AI model to search for and efficiently provide relevant information based on search keywords entered by users. The system operates through a process of analyzing the intent of the search keywords and retrieving relevant information from a document database.

[1692] System Overview

[1693] The system sends search keywords entered by users to a server, which then uses a generative AI model to analyze the intent of the keywords, retrieving relevant information from a database of materials and providing it to the user.

[1694] Hardware or software used

[1695] The following hardware and software are used to implement this system:

[1696] Device: The device on which a user enters search keywords. Examples include PCs, smartphones, and tablets.

[1697] Server: A central processing unit that analyzes search keywords, searches the document database, and generates results. It can be a cloud server or an on-premise server.

[1698] Generative AI model: An artificial intelligence algorithm that analyzes the intent of keywords and generates relevant information. For example, a language model such as GPT-3 is used.

[1699] Material database: A data store that stores the data to be searched. It can be an SQL database, a NoSQL database, ElasticSearch, etc.

[1700] Explaining program processing in natural language

[1701] User search keyword input

[1702] The user enters any keyword into the search bar of the terminal. For example, they enter "project progress management template." When the user clicks the search button, the terminal constructs this keyword and prepares to send it to the server.

[1703] Sending keywords to the server

[1704] The device constructs a JSON-formatted request from the entered keywords and sends this request to the server. The JSON format ensures that the keywords are sent to the server in a properly formatted state.

[1705] Analyze keyword intent

[1706] The server passes the received search keywords to the generative AI model, which analyzes the intent of the keywords and infers related contexts and phrases. For example, for "project progress management template," it generates related keywords such as "project management," "progress status," and "progress report."

[1707] Search the document database

[1708] The server uses the relevant keywords obtained from the generative AI model to search a database of documents, which contains various documents, to identify documents that match or are related to the relevant keywords. The search can use partial match or full-text search techniques.

[1709] Generating search results

[1710] The server lists relevant documents retrieved from the resource database and generates search results to provide relevant information to the user, including document titles, summaries, and links.

[1711] Sending search results to the server

[1712] The server then sends the generated search results to the device as a JSON response, which includes detailed information about related documents for easy user access.

[1713] Displaying search results

[1714] The terminal analyzes the search results received from the server and displays them on the user interface. The terminal presents the user with a list of related documents in an easy-to-understand format, allowing the user to quickly access the information they need.

[1715] Specific examples

[1716] For example, if the user inputs "template for project progress management," the system performs the following operations.

[1717] 1. The user enters search keywords into the device and clicks the search button.

[1718] 2. The device sends the keyword in JSON format to the server.

[1719] 3. The server passes the keywords to a generative AI model for analysis, generating related keywords such as "project management," "progress status," and "progress report."

[1720] 4. The server searches the resource database to identify relevant documents.

[1721] 5. The server generates a list of search results and sends them to the device in JSON format.

[1722] 6. The device analyzes the search results and displays them in the user interface.

[1723] In this way, the system of the present invention can accurately understand the intent of the keywords entered by the user and quickly provide related information. By analyzing specific phrases such as "project progress management template" as prompts, related information can be accurately obtained. This process improves the user's search efficiency and streamlines work.

[1724] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1725] Step 1:

[1726] The user inputs a search keyword. Specifically, the user inputs any keyword into the search bar of the terminal. As an example of input, the user inputs "template for project progress management." In this way, keywords are collected from the user.

[1727] Step 2:

[1728] The device constructs a request using the keywords in JSON format, for example in JSON format like this:

[1729] json

[1730] {

[1731] "search_query": "project progress management template"

[1732] }

[1733] The terminal constructs this JSON request as an HTTP POST request and sends it to the server. The input is the keyword entered by the user, and the output is the JSON-formatted request.

[1734] Step 3:

[1735] The server parses the received JSON request. The server extracts the value of "search_query" from the request and uses this value in the next step. The input is the JSON formatted request, and the output is the extracted keywords.

[1736] Step 4:

[1737] The server passes the keywords to a generative AI model, which analyzes their intent. The generative AI model is accessed through a specific API. For example, an API request can be used to send a "project progress management template" to the model and retrieve related phrases. The input is the extracted keywords, and the output is a list of related keywords.

[1738] Step 5:

[1739] The server searches the document database based on the related keywords obtained from the generative AI model. For example, it generates a database query using "project management," "progress status," "progress report," etc., and searches the document database. The input is the generated related keywords, and the output is a list of related documents.

[1740] Step 6:

[1741] The server generates a JSON formatted search result list to provide relevant information to the user, for example:

[1742] json

[1743] {

[1744] "results": [

[1745] {

[1746] "title": "Project Management Basics",

[1747] "summary": "Importance of project management and basic techniques",

[1748] "link": "http: / / example.com / doc1"

[1749] },

[1750] {

[1751] "title": "Progress Status Check Template",

[1752] "summary": "Templates for efficiently reviewing and reporting progress",

[1753] "link": "http: / / example.com / doc2"

[1754] }

[1755] ]

[1756] }

[1757] The input is a list of relevant documents, and the output is the search results in JSON format.

[1758] Step 7:

[1759] The server generates search results and sends them to the terminal as a JSON response. The input is the JSON search results and the output is the HTTP response.

[1760] Step 8:

[1761] The terminal analyzes the search results received from the server and displays them on the user interface. Specifically, it analyzes the JSON data of the search results, generates HTML and CSS, and displays them to the user. The input is the JSON-formatted search results, and the output is the information displayed on the user interface.

[1762] Through the above processing steps, a system is realized that accurately understands the intent of the keywords entered by the user and quickly provides related information.

[1763] (Application example 1)

[1764] 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."

[1765] Conventional search systems have the drawback of being inefficient, requiring users to manually search through a vast amount of information to find the data they need. Furthermore, they often fail to accurately understand the intent of search keywords, resulting in irrelevant results being returned. This results in a poor user experience and wastes time and effort.

[1766] 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.

[1767] In this invention, the server includes means for receiving search keywords entered by the user, means for passing the received search keywords to a generative AI and analyzing the intent, means for searching a document database based on related keywords obtained from the generative AI, means for providing search results to the user, and means for controlling the display of search results on the user terminal, thereby enabling the information required by the user to be provided quickly and accurately.

[1768] The "means for receiving search keywords input by a user" is a function that allows the system to receive search keywords input by a user to a terminal.

[1769] "Means for passing received search keywords to a generative AI to analyze the intent" is a function for passing received search keywords to a generative AI and analyzing the intent and purpose of the keywords.

[1770] "Means for searching the document database based on related keywords obtained from generative artificial intelligence" is a function for searching the document database using related keywords extracted from the results of analysis by generative artificial intelligence.

[1771] The "means for providing search results to the user" is a function for presenting search results obtained from the document database to the user.

[1772] The "means for controlling the display of search results on the user terminal" is a function for appropriately displaying search results on the user terminal.

[1773] "Means for sending received search keywords to the server in JSON format" is a function for converting search keywords received from a user into JSON format and sending them to the server.

[1774] "Generative AI infers the user's intent from the search keywords and generates related keywords" is a function in which generative AI infers the user's purpose based on the search keywords and derives highly relevant keywords.

[1775] "Sends related keywords obtained from generative AI to a document database to search for related information, formats the search results, and displays them on a user interface" is a function that searches a document database using related keywords obtained from generative AI, and displays the results in a format that is easy for the user to understand.

[1776] This invention relates to a system that uses generative artificial intelligence to search for and efficiently provide relevant information based on search keywords entered by users. The system operates through a process of analyzing the intent of the search keywords and retrieving relevant information from a document database.

[1777] System Overview

[1778] The system includes the following elements:

[1779] 1. A means of receiving search keywords entered by users

[1780] 2. A method for passing received search keywords to a generative AI to analyze intent

[1781] 3. A method for searching a database of materials based on related keywords obtained from generative artificial intelligence

[1782] 4. How search results are presented to users

[1783] 5. Means of controlling the display of search results on the user's device

[1784] System Operation

[1785] User search keyword input

[1786] The user enters a keyword into the search bar of the device, for example, "benefit list." When the user clicks the search button, the device constructs this keyword and prepares to send it to the server.

[1787] Sending keywords to the server

[1788] The device constructs a JSON-formatted request from the entered keywords and sends this request to the server. The JSON-formatted request ensures that the keywords are sent to the server in a properly formatted state.

[1789] Analyze keyword intent

[1790] The server passes the received search keywords to a generative AI, which analyzes the intent of the keywords and infers related contexts and phrases. For example, for a "benefit list," it generates related keywords such as "benefits," "coupons," and "point redemption."

[1791] Search the document database

[1792] The server uses the related keywords obtained from the generative artificial intelligence to search the resource database, which contains a variety of information, and identifies information that matches or is related to these related keywords. The search can use partial match or full-text search techniques.

[1793] Generating search results

[1794] The server lists related information retrieved from the document database and generates search results for providing the related information to the user, including the title, summary, and link of the information.

[1795] Sending search results to the server

[1796] The server generates search results and sends them to the device as a JSON response, which includes relevant details and makes them easily accessible to the user.

[1797] Displaying search results

[1798] The terminal analyzes the search results received from the server and displays them on the user interface. The terminal presents the user with a list of related information in an easy-to-understand format, allowing the user to quickly access the information they need.

[1799] Specific examples

[1800] For example, if the user types "benefit list," the system will do the following:

[1801] 1. The user enters search keywords into the device and clicks the search button.

[1802] 2. The device sends the keyword in JSON format to the server.

[1803] 3. The server passes the keywords to a generative AI system for analysis, which generates related keywords such as "benefits," "coupons," and "point redemption."

[1804] 4. The server searches the resource database to identify relevant information.

[1805] 5. The server generates a list of search results and sends them to the device in JSON format.

[1806] 6. The device analyzes the search results and displays them in the user interface.

[1807] Example prompts to input to the generative AI model

[1808] Benefits List

[1809] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1810] Step 1:

[1811] The user enters a search keyword into the device's search bar. For example, they enter "benefit list." The entered keyword "benefit list" is temporarily stored in the device's internal memory. When the user clicks the search button, the device begins preparing to send the keyword.

[1812] Step 2:

[1813] The device converts the input keyword "Benefits List" into a JSON request. Specifically, it formats the keyword appropriately and generates the following JSON data:

[1814] json

[1815] {

[1816] "keyword": "Benefits list"

[1817] }

[1818] This JSON formatted request is sent to the server as an HTTP POST request, which allows the server to understand what the user wants to search for.

[1819] Step 3:

[1820] The server analyzes the received JSON data and obtains the search keyword "Benefit List." The server passes this keyword to the generative AI and begins analysis. Specifically, the server inputs the following prompt into the generative AI:

[1821] Benefits List

[1822] Based on this prompt, the generative artificial intelligence analyzes the intent of the keywords and generates related keywords such as "benefits," "coupons," and "point redemption."

[1823] Step 4:

[1824] The server searches the document database based on the related keywords obtained from the generative AI. Specifically, it sends the following search query to the document database:

[1825] json

[1826] {

[1827] "related_keywords": ["benefits", "coupon", "points redemption"]

[1828] }

[1829] The resource database receives this query and searches for relevant information. The server retrieves the list of identified information as search results and formats it.

[1830] Step 5:

[1831] The server generates a list of search results and sends it to the device as a JSON response, which includes a title, summary, and link. An example of a specific response is as follows:

[1832] json

[1833] {

[1834] "results": [

[1835] {"title": "Benefits List", "summary": "Latest Benefits List", "link": "https: / / example.com / benefits"},

[1836] {"title": "Coupon Information", "summary": "Available Coupon Information", "link": "https: / / example.com / coupons"},

[1837] {"title": "Points redemption", "summary": "Detailed information about points redemption", "link": "https: / / example.com / points"}

[1838] ]

[1839] }

[1840] Step 6:

[1841] The device parses the received JSON format search results and displays them on the user interface. Specifically, the parsed search results are displayed as follows:

[1842] 1. Benefits List: Latest Benefits List

[1843] Link: https: / / example.com / benefits

[1844] 2. Coupon Information: Available coupon information

[1845] Link: https: / / example.com / coupons

[1846] 3. Point Redemption: Detailed information on point redemption

[1847] Link: https: / / example.com / points

[1848] The user can view the results and quickly access the information they need.

[1849] Through the above processing steps, related information generated based on the search keyword entered by the user is efficiently presented.

[1850] 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.

[1851] ---

[1852] The present invention relates to a system that uses generative artificial intelligence (AI) to search for and efficiently provide relevant information based on search keywords entered by users. The present invention further improves search accuracy and user experience by combining it with an emotion engine that recognizes user emotions.

[1853] System Overview

[1854] In this system, search keywords entered by the user on their device are sent to a server, which then uses generative artificial intelligence and an emotion engine to analyze the intent of the keywords and the user's emotions. Based on the analysis results, related information is retrieved from a document database and provided to the user. The specific process flow and operation are explained below.

[1855] Explaining program processing in natural language

[1856] User search keyword input

[1857] The user inputs any keyword into the search bar of the terminal. For example, the user inputs "template for project progress management." When the user clicks the search button, the terminal constructs the keyword and prepares to send it to the server.

[1858] Send keywords and emotional information to the server

[1859] The device constructs a search request in JSON format that includes the entered keywords and emotional information obtained from the user's facial expressions and voice, and sends this request to the server. For example, the structure may look like the following:

[1860] {

[1861] "search_keyword": "Project progress management template",

[1862] "emotion_data": {

[1863] "happiness": 0.8,

[1864] "sadness": 0.1,

[1865] "anger": 0.1

[1866] }

[1867] }

[1868] Keyword and sentiment analysis

[1869] The server analyzes the received search keywords and emotional information. The server then passes the search keywords to a generative artificial intelligence (AI) that analyzes their intent. The emotion engine then analyzes the user's emotional information and adjusts related keywords based on that emotion. For example, if the user indicates a high level of happiness, it will prioritize generating positive related keywords.

[1870] Search the document database

[1871] The server uses the related keywords obtained from the generative artificial intelligence and emotion engine to search the document database, which contains various internal company documents, and identifies documents that match or are related to these related keywords. The search can use partial match and full-text search techniques.

[1872] Generating search results

[1873] The server retrieves search results from the resource database and generates a list of relevant documents. Each document contains information such as a title, summary, and links. For example, the search results look like this:

[1874] {

[1875] "results": [

[1876] {

[1877] "title": "Project Management Template",

[1878] "summary": "Project management template.",

[1879] "link": " / documents / project_management_template.pdf"

[1880] },

[1881] {

[1882] "title": "Progress report example",

[1883] "summary": "Template for reporting project progress.",

[1884] "link": " / documents / progress_report_example.pdf"

[1885] }

[1886] ]

[1887] }

[1888] Sending search results to the server

[1889] The server then sends the generated search results to the device as a JSON response. The search results are adjusted to prioritize relevant documents based on emotional information.

[1890] Displaying search results

[1891] The device receives the response from the server and analyzes the search results. The analysis results are displayed in the user interface. The user can view the titles, summaries, and links of related documents and access the information they need. Recommendations based on sentiment may also be displayed.

[1892] Specific examples

[1893] For example, if a user inputs "project progress management template" and simultaneously indicates a high happiness level, the system will perform the following actions:

[1894] 1. The user enters search keywords into the device and clicks the search button.

[1895] 2. The device sends keywords and emotion information in JSON format to the server.

[1896] 3. The server passes the keywords to the generative AI and the emotional information to the emotion engine for analysis.

[1897] 4. Generative AI generates related keywords such as "project management," "progress status," and "progress report."

[1898] 5. The emotion engine prioritizes positive related keywords based on high happiness.

[1899] 6. The server searches the resource database to identify relevant documents.

[1900] 7. The server generates a list of search results and sends them to the device in JSON format.

[1901] 8. The device analyzes the search results and displays them in the user interface.

[1902] In this way, the system of the present invention can accurately grasp the user's intentions and emotions and provide relevant information quickly and appropriately, thereby improving the user's search experience and increasing work efficiency.

[1903] The processing flow will be explained below.

[1904] ---

[1905] Step 1:

[1906] The user enters "Project progress management template" in the search bar of the device. The user clicks the search button.

[1907] Step 2:

[1908] The device detects user input and constructs a search request in JSON format that includes the entered keywords. At the same time, it collects emotional information obtained from the user's facial expressions and voice and adds it to the request. An example structure would be as follows:

[1909] {

[1910] "search_keyword": "Project progress management template",

[1911] "emotion_data": {

[1912] "happiness": 0.8,

[1913] "sadness": 0.1,

[1914] "anger": 0.1

[1915] }

[1916] }

[1917] Step 3:

[1918] The device sends the constructed search request to the server, using an HTTP POST request.

[1919] Step 4:

[1920] The server analyzes the search request received from the device and obtains search keywords and emotion information. For example, the server obtains the keyword "project progress management template" and emotion information.

[1921] Step 5:

[1922] The server calls a generative artificial intelligence (AI) and passes the acquired search keywords to the AI. The AI ​​analyzes the user's intent based on these keywords and generates related keywords. For example, the AI ​​generates the following related keywords:

[1923] [

[1924] "project management",

[1925] "Progress Status",

[1926] "Progress Report"

[1927] ]

[1928] Step 6:

[1929] The server calls the emotion engine and passes the acquired emotion information to the emotion engine. The emotion engine analyzes the user's emotion and adjusts the generated related keywords based on the user's happiness level. For example, if the user indicates a high happiness level, it will prioritize positive related keywords.

[1930] Step 7:

[1931] The server retrieves the adjusted related keywords obtained from the generative artificial intelligence and emotion engine, and uses them to search a document database, which stores business-related document data, using, for example, SQL queries or a full-text search engine.

[1932] Step 8:

[1933] The server retrieves search results from the resource database and generates a list of relevant documents. Each document contains information such as a title, summary, and links. For example, the search results look like this:

[1934] {

[1935] "results": [

[1936] {

[1937] "title": "Project Management Template",

[1938] "summary": "Project management template.",

[1939] "link": " / documents / project_management_template.pdf"

[1940] },

[1941] {

[1942] "title": "Progress report example",

[1943] "summary": "Template for reporting project progress.",

[1944] "link": " / documents / progress_report_example.pdf"

[1945] }

[1946] ]

[1947] }

[1948] Step 9:

[1949] The server generates search results and sends them to the terminal as a JSON response. The search results are adjusted to prioritize relevant documents based on emotional information.

[1950] Step 10:

[1951] The device receives the response from the server and analyzes the search results. The analysis results are displayed in the user interface. The user can view the titles, summaries, and links of related documents and access the information they need. Recommendations based on sentiment may also be displayed.

[1952] ---

[1953] Through these processing steps, users can efficiently find the information they need. This system accurately understands the user's intentions and emotions and provides relevant information quickly and appropriately, improving the efficiency of the entire search process and the user experience.

[1954] Example 2

[1955] 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."

[1956] Conventional information retrieval systems have difficulty accurately grasping the user's search intent, and the search results often do not meet the user's expectations. Furthermore, they provide search results without considering the user's emotions, which results in a poor user experience. It is necessary to solve these problems and provide highly accurate search results that take the user's intent and emotions into account.

[1957] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for acquiring search keywords and emotion information entered by a user in JSON format, means for transmitting the acquired search keywords and emotion information to the server, means for passing the received search keywords to a generative AI model to analyze the intention, means for passing the received emotion information to an emotion analysis engine to evaluate it, means for adjusting related keywords obtained from the generative AI model based on the results of the emotion analysis, means for searching a document database based on the adjusted related keywords, and means for providing search results to the user in JSON format. This makes it possible to provide highly accurate search results that accurately reflect the user's intention and emotion.

[1958] 1. "Search Keywords" are words or phrases that users enter to retrieve information they wish to obtain.

[1959] 2. "Emotional information" is data that represents the user's emotional state obtained from facial expressions, voice, etc.

[1960] 3. "JSON format" is a data format that structures data in text format to make it easier to read.

[1961] 4. "Server" refers to a computer system that receives a request from a client, processes it, and returns a response.

[1962] 5. A "generative AI model" is an artificial intelligence system that generates relevant information or keywords based on specified input.

[1963] 6. "Intent analysis" is the process of understanding the goals and needs behind a user's search keywords.

[1964] 7. "Emotion analysis engine" is a system for analyzing a user's emotional information and evaluating its content.

[1965] 8. “Related Keywords” are additional terms related to the user’s intent that are generated by the generative AI model based on the search keywords.

[1966] 9. "Document Database" means a database containing various information and documents, intended to provide information in response to a search query.

[1967] 10. "Search Results" means a list of information related to the user's search keywords retrieved from the resource database.

[1968] 11. The "HTTP POST method" is the part of the HTTP protocol that allows a client to send data to a server.

[1969] 12. "User interface" means the interface through which a user interacts with a system.

[1970] The present invention relates to a search system that uses a generative AI model and a sentiment analysis engine to search for and efficiently provide relevant information based on search keywords entered by users. The present invention also improves search accuracy and user experience by taking user sentiment into account.

[1971] System Overview

[1972] This system sends search keywords and emotional information entered by users to a server, which then uses a generative AI model and an emotional analysis engine to analyze the intent of the keywords and the user's emotions. Based on the analysis results, it retrieves related information from a document database and provides it to the user.

[1973] Hardware and Software Configuration

[1974] Terminal

[1975] The terminal is a computing device such as a PC, smartphone, or tablet, and provides a means for inputting search keywords and emotion information using a user interface.

[1976] server

[1977] The server is a computer system that receives and processes requests, analyzes search keywords and sentiment information, generates related information, searches the database, and provides results. The following software can be used:

[1978] Generative AI models: Use natural language processing models (e.g., GPT-4) to analyze user intent and generate relevant keywords.

[1979] Sentiment analysis engine: Evaluates user emotional information using sentiment analysis technology (e.g., IBM Watson Emotion Analysis).

[1980] Document database: Document searches are performed using full-text search engines such as Elasticsearch.

[1981] Specific program description

[1982] Input of user search keywords and emotional information

[1983] The user types "project progress management template" into the device's search bar, and the device simultaneously captures emotional information from the user's facial expressions and voice via the device's camera and microphone.

[1984] Sending keywords and sentiment information

[1985] The device constructs the acquired search keywords and emotion information in JSON format and sends it to the server using the HTTP POST method.

[1986] Keyword and emotional analysis

[1987] The server analyzes the received JSON-formatted data and passes the search keywords to the AI ​​model for generation and the emotional information to the emotion analysis engine.

[1988] Generate and adjust related keywords

[1989] The generative AI model generates related keywords based on search keywords. It then adjusts the generated related keywords based on the output of the sentiment analysis engine to create appropriate keywords based on the user's emotional state. For example, if a high level of happiness is indicated, positive keywords will be prioritized.

[1990] Search the document database

[1991] The server uses the tailored related keywords to search a document database, which contains internal and public documents, to identify relevant documents based on the search query.

[1992] Generating and serving search results

[1993] The server creates a JSON-formatted response with the search results and sends it to the device. The device then parses the results and displays them in a user interface. The user can then view the titles, summaries, and links of related documents and access the information they need.

[1994] Specific examples

[1995] For example, if a user types "project progress management template," the following prompt is sent to the generative AI model:

[1996] "Project Progress Management Template"

[1997] If the emotional information is high in happiness, the generated related keywords include "successful project management examples" and "positive progress reports," providing search results that meet user expectations and resulting in a more satisfying search experience.

[1998] In this way, the system of the present invention provides highly accurate search results that reflect the user's intentions and emotions, improving the search experience and increasing work efficiency.

[1999] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2000] Step 1:

[2001] The user inputs search keywords and emotional information. The user types "project progress management template" into the search bar on their device, and at the same time, emotional information (facial expressions and voice) is acquired via the camera and microphone. The input data obtained is the search keywords in text form and emotional information in numerical form (happiness, sadness, anger, etc.). This becomes the input for the next processing step.

[2002] Step 2:

[2003] The device sends keywords and emotion information to the server. The device then constructs the acquired search keywords and emotion information into a JSON-formatted request and sends it to the server using the HTTP POST method. The input is the search keywords and emotion information, and the output is the JSON-formatted request data.

[2004] Step 3:

[2005] The server analyzes keywords and emotional information. The server first analyzes the received JSON data and separates the search keywords from the emotional information. The server passes the search keywords to the generative AI model and the emotional information to the emotional analysis engine. The input is JSON data, and the output is the provision of individual data to the generative AI model and the emotional analysis engine.

[2006] Step 4:

[2007] The server generates and adjusts related keywords. The generative AI model analyzes search keywords and generates related keywords. Furthermore, the sentiment analysis engine evaluates emotional information and adjusts the generated related keywords. For example, if a high level of happiness is indicated, more positive related keywords are prioritized. The input is data provided to the generative AI model and sentiment analysis engine, and the output is a list of adjusted related keywords.

[2008] Step 5:

[2009] The server searches the document database. Based on the adjusted related keywords, the server queries the document database (e.g., Elasticsearch) to find the relevant documents. The input is the list of adjusted related keywords, and the output is a list of search results.

[2010] Step 6:

[2011] The server generates search results and sends them to the terminal. The server reconstructs the search results obtained from the database into JSON format and sends them to the terminal as an HTTP response. The input is a list of search results, and the output is the response data in JSON format.

[2012] Step 7:

[2013] The terminal displays the search results. The terminal receives the response data from the server, analyzes the search results, and displays them in the user interface. The user can check the titles, summaries, and links of related documents to access the information they need. The input is the response data in JSON format, and the output is the displayed search results.

[2014] In this way, highly accurate search results that reflect the user's intentions and emotions are provided.

[2015] (Application example 2)

[2016] 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."

[2017] Conventional search systems provide search results without considering the user's emotions, which results in poor user experience. Furthermore, they are unable to accurately grasp the user's search intent, making it difficult to provide appropriate and relevant information. Therefore, there is a need for a system that can adjust search results based on the user's emotions and provide more appropriate and relevant information.

[2018] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving search keywords entered by the user, means for transmitting the received search keywords and the user's emotion data to the server in JSON format, means for passing the search keywords and the user's emotion data to a generative AI to analyze the intention and generate related keywords, means for analyzing the user's emotion using an emotion engine and adjusting the related keywords based on the emotion, means for searching a document database, and means for providing the search results to the user. This makes it possible to provide appropriate and highly relevant information that takes the user's emotion into consideration.

[2019] The "means for receiving search keywords input by the user" is an interface for obtaining search keywords input by the user.

[2020] "Means of passing received search keywords to generative artificial intelligence (AI) to analyze intent" refers to the process of sending acquired search keywords to generative artificial intelligence (AI) and analyzing their intent.

[2021] "Means for sending the received search keywords and user emotion data to the server in JSON format" is a function for converting the acquired search keywords and user emotion data into JSON format and sending it to the server.

[2022] "Means for searching a document database based on related keywords obtained from generative artificial intelligence" refers to the process of searching the contents of a document database using related keywords generated by generative artificial intelligence.

[2023] "Means for analyzing user emotions using an emotion engine and adjusting related keywords based on those emotions" refers to a process for analyzing user emotional information using an emotion engine and correcting and adjusting the generated related keywords based on the results.

[2024] "Means for providing search results to users" refers to the interface for organizing the search results and displaying and providing them to users.

[2025] A "document database" is a searchable database in which documents and information are stored.

[2026] This invention is a system that optimizes search results by combining generative artificial intelligence and an emotion engine based on search keywords and emotion data entered by the user.

[2027] System Overview

[2028] Users input search keywords and emotion data using a device (smartphone or head-mounted display). The system sends this data to a server, which then uses generative artificial intelligence and an emotion engine to analyze the user's search intent and emotion. The system then searches for relevant information from a database and provides it to the user.

[2029] Hardware and software used

[2030] Hardware: Smartphone, Head-Mounted Display (HMD)

[2031] Software: Generative AI, emotion engine, JSON processing library (e.g., Python's requests library)

[2032] Operating procedure

[2033] User Input

[2034] The user enters search keywords using the device interface. For example, the user searches for "summer dress." At the same time, the user's emotional data (e.g., happiness, sadness, anger) is also acquired.

[2035] Sending data to the server

[2036] The device converts the search keywords and emotion data into JSON format and sends it to the server as follows:

[2037] {

[2038] "search_keyword": "summer dress",

[2039] "emotion_data": {

[2040] "happiness": 0.9,

[2041] "sadness": 0.0,

[2042] "anger": 0.1

[2043] }

[2044] }

[2045] Intention and emotion analysis

[2046] The server passes the received search keywords and emotion data to a generative AI to analyze the user's intention. The emotion engine also analyzes the user's emotions, and if the emotion is highly positive, it adjusts the related keywords in a more positive direction.

[2047] Search the document database

[2048] Based on the generated related keywords, the server searches the document database for related information such as "summer dress" or "summer dress."

[2049] Providing search results

[2050] The server sends the search results as a JSON response to the terminal, which receives the response and displays it in its user interface.

[2051] Specific examples

[2052] For example, if a user inputs the keyword "summer dress" and the emotional data indicates a high happiness level, the system operates as follows:

[2053] 1. A user enters "summer dress" and the emotion data is "happiness: 0.9, sadness: 0.0, anger: 0.1".

[2054] 2. The device converts this into JSON format and sends it to the server.

[2055] 3. The server passes the received data to the generative AI and emotion engine for analysis.

[2056] 4. Based on the emotional data, the generative AI will prioritize keywords that are related to positive products when you are in a "happy mood."

[2057] 5. The server searches the relevant database of materials to obtain information on positive products.

[2058] 6. The server sends the search results to the device in JSON format.

[2059] 7. The device analyzes the results and displays them in the user interface.

[2060] Prompt Sentence Examples

[2061] "If a user searches for 'summer dress' and the sentiment data looks like this:

[2062] Happiness: 0.9

[2063] Sadness: 0.0

[2064] Anger: 0.1

[2065] In response to this, please propose related products with vibrant and colorful designs.

[2066] This system allows users to get optimal search results tailored to their emotional state, greatly improving the search experience.

[2067] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2068] Step 1:

[2069] The user enters search keywords using the device's interface. For example, the user enters "summer dress." At this time, the device acquires the user's emotional data (happiness, sadness, anger, etc.) along with the search keywords. The inputs are the search keywords in text format and sensor data (emotional data).

[2070] Step 2:

[2071] The device converts the acquired search keywords and emotion data into JSON format and constructs a JSON object like the one below.

[2072] {

[2073] "search_keyword": "summer dress",

[2074] "emotion_data": {

[2075] "happiness": 0.9,

[2076] "sadness": 0.0,

[2077] "anger": 0.1

[2078] }

[2079] }

[2080] This JSON object is sent to the server, which gets the text data and emotion data as input and the constructed JSON object as output.

[2081] Step 3:

[2082] The server parses the received JSON object and extracts search keywords and sentiment data. Specifically, it breaks down the data through JSON parsing. The input is a JSON object, and the output is the extracted text data and sentiment data.

[2083] Step 4:

[2084] The server then passes the extracted search keywords to a generative artificial intelligence (AI) for analysis. This AI then infers the user's intent from the search keywords and executes a process to generate related keywords. The input is text data (search keywords), and the output is a list of related keywords.

[2085] Step 5:

[2086] The server analyzes the user's emotional data using an emotion engine. This emotion engine adjusts the list of related keywords based on the emotional data. The input is the emotional data and the list of related keywords, and the output is the adjusted list of keywords based on the emotions.

[2087] Step 6:

[2088] The server searches the database of materials based on the adjusted keyword list. It uses the adjusted keywords as a search query to retrieve matching or similar materials from the database. The input is the adjusted keyword list, and the output is a list of related materials.

[2089] Step 7:

[2090] The server organizes the acquired data in JSON format and constructs a response like the one below.

[2091] {

[2092] "results": [

[2093] {

[2094] "title": "Summer dress - colorful design",

[2095] "summary": "A colorful and gorgeous summer dress.",

[2096] "link": " / products / summer_dress_1"

[2097] },

[2098] {

[2099] "title": "Summer dress - cool material",

[2100] "summary": "A summer dress made from a refreshing material.",

[2101] "link": " / products / summer_dress_2"

[2102] }

[2103] ]

[2104] }

[2105] The input is a list of documents, and the output is a constructed JSON response.

[2106] Step 8:

[2107] The terminal receives the JSON response from the server, parses it, and displays it in the user interface, including the title, summary, and link of the search result. The input is the JSON response, and the output is the search result displayed on the screen.

[2108] This way, users receive relevant search results based on their emotions.

[2109] 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.

[2110] 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.

[2111] 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.

[2112] 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.

[2113] 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.

[2114] 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.

[2115] 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).

[2116] 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.

[2117] 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."

[2118] 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.

[2119] 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).

[2120] 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.

[2121] 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.

[2122] 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.

[2123] 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.

[2124] 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.

[2125] 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.

[2126] 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.

[2127] 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.

[2128] 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.

[2129] 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.

[2130] The following is further disclosed regarding the above embodiment.

[2131] ---

[2132] (Claim 1)

[2133] means for receiving search keywords entered by a user;

[2134] A means of passing received search keywords to a generative AI to analyze intent,

[2135] A means for searching a database of materials based on related keywords obtained from generative artificial intelligence;

[2136] a means for providing search results to a user;

[2137] A system including:

[2138] (Claim 2)

[2139] 10. The system of claim 1, further comprising: means for transmitting the received search keywords to the server in JSON format.

[2140] (Claim 3)

[2141] The system of claim 1, wherein the generative artificial intelligence infers the user's intention from the search keywords and generates related keywords.

[2142] "Example 1"

[2143] (Claim 1)

[2144] means for receiving search keywords entered by a user;

[2145] A means of passing received search keywords to a generative AI model to analyze intent,

[2146] A means of searching the document database based on relevant keywords obtained from the generative AI model; and

[2147] a means for providing search results to a user;

[2148] A means for the terminal to construct a JSON format request from the input keyword and send it to the server;

[2149] A means for the server to send search results to the terminal as a JSON format response;

[2150] A system including:

[2151] (Claim 2)

[2152] 10. The system of claim 1, further comprising means for analyzing search results received by the terminal and displaying them on a user interface.

[2153] (Claim 3)

[2154] The system of claim 1, wherein the generative AI model infers the user's intent from the search keywords and generates related keywords.

[2155] "Application Example 1"

[2156] (Claim 1)

[2157] means for receiving search keywords entered by a user;

[2158] A means of passing received search keywords to a generative AI to analyze intent,

[2159] A means for searching a database of materials based on related keywords obtained from generative artificial intelligence;

[2160] a means for providing search results to a user;

[2161] means for controlling the display of search results on a user terminal;

[2162] A system including:

[2163] (Claim 2)

[2164] 10. The system of claim 1, further comprising: means for transmitting the received search keywords to the server in JSON format.

[2165] (Claim 3)

[2166] The system of claim 1, wherein the generative artificial intelligence infers the user's intention from the search keywords and generates related keywords.

[2167] (Claim 4)

[2168] The system according to claim 1, wherein the related keywords obtained from the generative artificial intelligence are sent to a document database to search for related information, and the obtained search results are formatted and displayed on a user interface.

[2169] "Example 2: Combining Emotion Engines"

[2170] (Claim 1)

[2171] A means to obtain the search keywords and sentiment information entered by the user in JSON format,

[2172] means for transmitting the acquired search keywords and emotion information to a server;

[2173] A means for the server to pass the received search keywords to a generative AI model to analyze the intent;

[2174] A means for passing the emotion information received by the server to an emotion analysis engine for evaluation;

[2175] A means for adjusting the related keywords obtained from the generative AI model based on the results of sentiment analysis;

[2176] a means of searching the resource database based on tailored related keywords;

[2177] A means of providing search results to users in JSON format;

[2178] A system including:

[2179] (Claim 2)

[2180] 2. The system according to claim 1, further comprising means for transmitting the received search keywords and emotion information to the server using an HTTP POST method.

[2181] (Claim 3)

[2182] The system of claim 1, wherein the generative AI model includes means for inferring user intent from search keywords and adjusting related keywords based on emotional information.

[2183] "Application example 2 when combining emotion engines"

[2184] (Claim 1)

[2185] means for receiving search keywords entered by a user;

[2186] A means of passing received search keywords to a generative AI to analyze intent,

[2187] A means to send the received search keywords and user emotion data to the server in JSON format;

[2188] A means for searching a database of materials based on related keywords obtained from generative artificial intelligence;

[2189] a means for analyzing a user's emotions using an emotion engine and adjusting related keywords based on the emotions;

[2190] a means for providing search results to a user;

[2191] A system including:

[2192] (Claim 2)

[2193] 10. The system of claim 1, further comprising means for transmitting the received search keywords and emotion data to the server in JSON format.

[2194] (Claim 3)

[2195] The system according to claim 1, wherein the generative artificial intelligence infers the user's intention from the search keywords, generates related keywords, and the emotion engine adjusts the related keywords based on the emotion information. [Explanation of symbols]

[2196] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving search keywords entered by a user; A means of passing received search keywords to a generative AI to analyze intent, A means for searching a database of materials based on related keywords obtained from generative artificial intelligence; a means for providing search results to a user; A system including:

2. The system of claim 1 , further comprising: means for transmitting the received search keywords to a server in JSON format.

3. 2. The system according to claim 1, wherein the generative artificial intelligence infers the user's intention from the search keywords and generates related keywords.

Images