system

Abstract

Provide a system. A method for receiving a search query includes: means for obtaining search results from a search engine based on the search query; means for transmitting the search query to an artificial intelligence model for generating the search query and obtaining the generated result; means for comparing the search results with the generated results and calculating the degree of match; A means for scoring the information accuracy of the search results based on the generated results; means for re-ranking the search results based on the degree of match and the information accuracy; means for displaying the re-ranked search results on a terminal; A system including:

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] Conventional search engines return many results based on a search query, but the reliability and accuracy of the results vary, making it difficult for users to quickly find the most relevant information. Even when generative AI is used, the generated results do not necessarily match all search results, so there is a need to efficiently provide the most reliable information to users. [Means for solving the problem]

[0005] The present invention provides a system including means for receiving a search query and obtaining search results from a search engine based on the search query, means for transmitting the search query to an AI model that generates the search query and obtaining the generated results, means for comparing the search results with the generated results and calculating a degree of match, means for scoring the information accuracy of the search results based on the generated results, means for re-ranking the search results based on the degree of match and the information accuracy, and means for displaying the re-ranked search results on a terminal, thereby enabling a user to obtain more reliable search results using the results of the AI ​​that generates the search results as a verification standard.

[0006] A "search query" is a piece of text or phrase that a user enters into a search engine to search for information.

[0007] A "search engine" is a software system that searches for information on the Internet based on specified keywords and presents the results to the user.

[0008] A "generative artificial intelligence model" is an artificial intelligence algorithm or model that can use natural language processing to generate answers or information relevant to an input search query.

[0009] "Search results" are the list of web pages or information that a search engine returns to a user based on a search query.

[0010] "Generated results" are answers or information generated by a generating artificial intelligence model in response to a search query.

[0011] The "degree of match" is a numerical score that indicates how closely each item in the search results matches the generated results.

[0012] "Information accuracy" is a score used to evaluate the accuracy and reliability of search results based on the generated results.

[0013] "Reranking" is the process of re-evaluating the ranking of search results based on relevance and information accuracy to determine new rankings.

[0014] A "terminal" is a device such as a computer, smartphone, or tablet that a user uses to view search results. [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] The present invention is a system for optimizing search results based on a search query entered by a user using a search engine and a generating artificial intelligence model. The system performs the following consecutive processes: receiving a search query, obtaining search results, obtaining generated results from the generating AI, comparing and scoring the results, re-ranking, and displaying the final results.

[0037] First, a user enters a search query into a device. This query is sent to a search engine on the Internet, and traditional search results are obtained. In parallel, the server sends the same search query to an artificial intelligence model that generates it, and receives the generated results from the AI.

[0038] Next, the server compares the acquired search results with the results generated by the AI. Specifically, it matches the titles and snippets of the search results with the corresponding parts of the generated results to calculate the degree of match. This degree of match serves as an index for evaluating the reliability of each search result. Furthermore, the server scores the information accuracy of each search result based on the generated results. This quantifies how closely the AI's results match the actual search results.

[0039] Based on these relevance and information accuracy scores, the server re-ranks all search results. For example, a search result with a relevance score of 90 and an information accuracy score of 95 will receive a higher overall score of 92.5. In this way, the re-ranked search results are determined.

[0040] Finally, the server sends the re-ranked search results to the device and displays them to the user. This allows users to quickly obtain information with high reliability and guaranteed accuracy. For example, if a user enters the query "capital of Japan," the server will evaluate "Tokyo" as having a high degree of match and information accuracy and display it at the top.

[0041] By utilizing the advanced natural language processing capabilities of the AI ​​model it generates, this system is able to provide more reliable search results than conventional search engines, allowing users to quickly and accurately obtain the information they need.

[0042] The processing flow will be explained below.

[0043] Step 1:

[0044] The user enters a search query into the device. The user clicks the "Search" button to submit the query.

[0045] Step 2:

[0046] The server receives the user's search query and logs the query.

[0047] Step 3:

[0048] The server sends the received search query to the search engine, and waits for a response from the search engine.

[0049] Step 4:

[0050] The search engine returns the search results to the server, which temporarily stores the received search results in its internal database.

[0051] Step 5:

[0052] The server sends the same search query to the AI ​​model, and the server waits for the AI ​​model to generate the results.

[0053] Step 6:

[0054] The generating AI model sends the generated results back to the server, which stores the received results in its internal database.

[0055] Step 7:

[0056] The search results obtained by the server are compared with the results generated by the AI. The comparison mainly focuses on the titles and snippets of the search results.

[0057] Step 8:

[0058] The server calculates a match score for each search result, which involves quantifying the match between the content of the search result and the generated result.

[0059] Step 9:

[0060] The server scores each search result based on its accuracy, which is based on how reliable the generated result is.

[0061] Step 10:

[0062] The server re-ranks all search results based on relevance and information accuracy, calculates new rankings, and forms a re-ranked search result list.

[0063] Step 11:

[0064] The server sends the re-ranked search result list to the terminal, and the terminal displays the received search results for the user to review.

[0065] Step 12:

[0066] The user checks the displayed search results and obtains the information they need. The user clicks on the search results to view the details.

[0067] Example 1

[0068] 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."

[0069] Modern information retrieval systems often provide irrelevant or unreliable information in response to user-entered search queries. This prevents users from quickly obtaining the information they need, reducing search efficiency. Furthermore, traditional information retrieval services lack the means to adequately evaluate the reliability of search results, exposing users to the risk of making decisions based on incorrect information.

[0070] 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.

[0071] In this invention, the server includes means for receiving a search query, means for obtaining search results from an information search service, and means for sending a query to a generating artificial intelligence model and obtaining the generated results, thereby making it possible to calculate the degree of match between the obtained search results and the generated results, score the information accuracy based on that, and provide the user with re-ranked search results.

[0072] A "search query" is a string of characters that a user enters to search for information.

[0073] An "information search service" is a system or engine for obtaining information on the Internet, generally referring to a search engine.

[0074] A "generative artificial intelligence model" is a machine learning model for generating text based on a given query.

[0075] The "generation result" is the text data that the generating artificial intelligence model outputs in response to the input query.

[0076] "Search results" refers to the relevant information returned by an information retrieval service in response to a search query.

[0077] The "degree of match" is a numerical value or index that evaluates the degree of match between the search result and the generated result.

[0078] "Information accuracy" is a numerical value or index that indicates how accurate and reliable the generated results are.

[0079] "Reranking" is the process of reevaluating and rearranging the ranking of search results based on relevance and information accuracy.

[0080] "Terminal" refers to the device on which a user enters a search query and displays results, including a PC or smartphone.

[0081] This invention is a system for optimizing search results based on a user-provided search query by utilizing an information search service and a generative AI model. The system performs the following consecutive processes: receiving a search query, obtaining search results, obtaining generated results from a generative AI model, comparing the results and calculating the degree of match, scoring the information accuracy, re-ranking, and displaying the final results.

[0082] First, a user inputs a search query into a terminal. For example, the user inputs the query "capital of Japan." This query is sent to an information search service (e.g., a general-purpose search engine) via the Internet. At the same time, the same query is sent to a server.

[0083] The server sends the received search query to a generative AI model (e.g., a natural language processing model). Specifically, OpenAI's GPT-3 (registered trademark) is used. The generative AI model generates text based on the query and returns the generated results to the server.

[0084] Next, the server compares the search results obtained from the information search service with the results generated by the AI. This comparison is performed by matching the titles and summaries of the search results with the corresponding parts of the generated results. The server then calculates the degree of match. The degree of match is a numerical value that evaluates the degree of match between the search results and the generated results.

[0085] Next, the server performs a scoring process. Based on the results generated by the AI, the server scores the accuracy of each search result. Information accuracy is a numerical value that indicates how accurate and reliable the generated results are.

[0086] Based on these relevance and accuracy scores, the server re-ranks all search results. For example, if a search result has a relevance score of 90 and an accuracy score of 95, the overall score will be calculated as 92.5. Search results with higher scores will ultimately be ranked higher.

[0087] Finally, the server sends the re-ranked search results to the terminal, which then displays the re-ranked search results to the user, allowing the user to quickly check highly reliable and accurate information.

[0088] For example, if a user enters the query "capital of Japan," the server compares the results obtained from the information search service and the generative AI model, and scores search results containing "Tokyo" with high matching and information accuracy. As a result of the re-ranking process, information related to "Tokyo" is displayed at the top.

[0089] An example prompt is:

[0090] Query: capital of japan

[0091] Explanation: What is the capital of Japan?

[0092] By implementing this invention, users can quickly and accurately obtain the information they need. By utilizing the advanced natural language processing capabilities of the generated artificial intelligence model, it is possible to provide more reliable search results than conventional information search services.

[0093] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0094] Step 1:

[0095] User enters a search query

[0096] A user enters a search query into a device. For example, the user enters "capital of Japan."

[0097] Input: Search query (e.g. "Capital of Japan")

[0098] Output: Search queries sent from the device to the internet

[0099] Step 2:

[0100] Device submitting search queries

[0101] The terminal transmits the search query entered by the user to the information search service via the Internet, and also transmits the same query to the server.

[0102] Input: User's search query

[0103] Output: Search query sent to information retrieval service and server

[0104] Step 3:

[0105] Server obtains search results from information search service

[0106] The server retrieves search results from the information search service based on the submitted search query, using an API to retrieve results from a search engine (e.g., a general-purpose search engine).

[0107] Input: Search results from information search services

[0108] Output: Search results saved on the server

[0109] Step 4:

[0110] Server sends queries to the generative AI model

[0111] The server sends the received search query to a generative AI model, such as a natural language processing model (e.g., OpenAI's GPT-3).

[0112] Input: Search query

[0113] Output: Request to the generative AI model

[0114] Step 5:

[0115] Obtaining generated results from the generative AI model by the server

[0116] The generative AI model generates text based on the search query and returns the results to the server, which receives the generated results.

[0117] Input: Response from a generative AI model

[0118] Output: Generated results stored on the server

[0119] Step 6:

[0120] Comparing server search results with generated results

[0121] The server compares the search results from the information retrieval service with the generated results from the generative AI model by matching the titles and summaries of the search results with the corresponding parts of the generated results.

[0122] Input: Search results and generated results

[0123] Data processing: Matching with text comparison algorithms

[0124] Output: Match score for each result

[0125] Step 7:

[0126] Server-based scoring of information accuracy

[0127] The server scores the information accuracy of each search result based on the results generated by the AI. Information accuracy is a numerical value that evaluates the accuracy and reliability of the generated results.

[0128] Input: Generated results

[0129] Data Computation: Information Accuracy Scoring Algorithm

[0130] Output: Information accuracy score for each search result

[0131] Step 8:

[0132] Server Re-ranking

[0133] The server re-ranks all search results based on their relevance and accuracy scores, with more reliable results being ranked higher.

[0134] Input: Match score and information accuracy score

[0135] Data arithmetic: Re-ranking by weighted average of scores

[0136] Output: Re-ranked search results

[0137] Step 9:

[0138] Server sends re-ranked search results

[0139] The server transmits the re-ranked search results to the terminal.

[0140] Input: Reranked search results

[0141] Output: Search results sent to your device

[0142] Step 10:

[0143] Displaying search results by device

[0144] The terminal displays the re-ranked search results to the user, who can then check the displayed search results and quickly and accurately obtain the information they need.

[0145] Input: Reranked search results

[0146] Output: Search results displayed to the user

[0147] (Application example 1)

[0148] 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."

[0149] In conventional food delivery services, even when a user inputs the name of a desired dish or restaurant, the accuracy and reliability of the search results are insufficient, making it difficult to quickly find the best option. Furthermore, the search results often contain advertisements and irrelevant information, hindering the user's ability to select the best restaurant. The objective of the present invention is to solve these problems and quickly and accurately provide the information desired by the user.

[0150] 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.

[0151] In this invention, the server includes means for receiving a search query, means for obtaining search results from a search engine based on the search query, means for transmitting the search query to an AI model that generates the search query and obtaining the generated results, means for comparing the search results with the generated results and calculating a degree of match, means for scoring the information accuracy of the search results based on the generated results, means for re-ranking the search results based on the degree of match and the information accuracy, means for displaying the re-ranked search results on a terminal, and means for presenting food delivery options optimized using the AI ​​model based on the name of a dish or restaurant entered by a user. This enables a user to quickly obtain more reliable and accurate food delivery information in response to a search query.

[0152] A "search query" is a keyword or phrase that a user enters when searching for information.

[0153] A "search engine" is a system that searches for information on the Internet and provides related web pages and data.

[0154] A "generative artificial intelligence model" is an AI algorithm that performs natural language processing based on user input and generates information.

[0155] The "degree of match" is an index for evaluating the degree of match between the search results and the generated results.

[0156] "Information accuracy" is a score that indicates the accuracy and reliability of the information calculated by the generative AI model.

[0157] "Re-ranking" is a process of re-ranking the acquired search results based on the scores of match and information accuracy.

[0158] A "terminal" is an electronic device such as a personal computer, smartphone, or tablet.

[0159] The "dish name" is the name of a specific food item that a user specifies when using a food delivery service.

[0160] "Restaurant name" refers to the name of the restaurant or cafe that provides the food delivery service.

[0161] "Food delivery options" refers to multiple delivery service options offered for the food or restaurant searched for by the user.

[0162] The present invention provides a system for optimizing search results for food delivery services based on a search query entered by a user. The system supports a series of processes including receiving a search query, obtaining search results, obtaining generated results using an artificial intelligence model, comparing and scoring both results, re-ranking, and displaying the final results.

[0163] Hardware and Software Configuration

[0164] Server: A server with powerful processing power is responsible for processing search queries and aggregating results.

[0165] Device: A device such as a personal computer, smartphone, or tablet is used as the device where the user enters the search query and displays the final results.

[0166] software:

[0167] Programming language: Python

[0168] Libraries: Requests library (used to communicate with search engine APIs and food delivery APIs)

[0169] Generative AI models: Leverage artificial intelligence generative models (e.g., GPT-3) through the OpenAI API.

[0170] Data processing and calculation flow

[0171] The server first receives a search query entered by a user on a device, sends the search query to a food delivery search engine, and retrieves relevant search results. In parallel, the same search query is sent to a generative artificial intelligence model, and retrieves generated results.

[0172] The server then compares the search engine results with the results generated by the generative AI model and calculates the degree of match between them. This degree of match is calculated based on the degree of match between the title and snippet of the search result and the title and snippet of the generated result. The information accuracy of the search result is also scored based on the reliability of the generated result.

[0173] The server re-ranks the search results based on the match and information accuracy scores, and the re-ranked search results are sent to the terminal and displayed to the user.

[0174] Specific examples

[0175] For example, if a user types "sushi" into their device, the server retrieves food delivery options related to "sushi" from the search engine and the generative AI model. Then, it compares both results and re-ranks them based on match and information accuracy. Finally, the user is presented with the most reliable and accurate sushi delivery options.

[0176] Prompt Sentence Examples

[0177] "Give me the best food delivery options for Sushi"

[0178] Such a system allows users to obtain reliable food delivery information more efficiently.

[0179] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0180] Step 1:

[0181] A user inputs a search query into a terminal. The input search query is the name of a dish or restaurant that the user is interested in, such as "sushi." The input query becomes the basis for other processing steps in the system.

[0182] Step 2:

[0183] The server receives a search query. The received search query is sent to a search engine API to retrieve relevant search results. For example, if the search query is "sushi," the server retrieves multiple food delivery options that offer sushi. The input is the search query "sushi," and the output is a list of relevant search results.

[0184] Step 3:

[0185] The server sends the same search query to a generative AI model. The generative AI model provides generated results based on the query, for example using the OpenAI API. The input is the search query "sushi," and the output is a list of recommended food delivery options provided by the generative AI. The prompt provided by the generative AI model is "Give me the best food delivery options for Sushi."

[0186] Step 4:

[0187] The server compares the search results with the results of the generative AI model. Specifically, it compares the titles and snippets of the search results with those of the generated results and calculates the degree of match. The degree of match is quantified, for example, to show how many common elements each result contains. The input is the search results and the generated results, and the output is a score indicating the degree of match between each result.

[0188] Step 5:

[0189] The server scores the information accuracy of the search results based on the generated results. In this process, an information accuracy score is assigned to each search result based on the reliability index of the generative AI model and the content of the search result. The input is the match score and the reliability index of the generative AI model, and the output is the information accuracy score for each search result.

[0190] Step 6:

[0191] The server re-ranks the search results based on the relevance and accuracy scores. This re-ranking adjusts the ranking so that results with high relevance and accuracy are displayed at the top. The input is the relevance and accuracy scores of each search result, and the output is a list of re-ranked search results.

[0192] Step 7:

[0193] The server sends the re-ranked search results to the device. The user reviews these results on the device and selects the food delivery option that best suits them. The input is a list of re-ranked search results, and the output is the optimized search results displayed on the device.

[0194] 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.

[0195] The present invention combines a system that utilizes a search engine and a generating artificial intelligence model to optimize search results based on a search query entered by a user, with an emotion engine that recognizes the user's emotions. The system performs the following continuous processes: receiving a search query, obtaining search results, obtaining generated results from the generating AI, comparing and scoring the results, re-ranking, displaying the final results, and filtering and adjusting the search results based on the user's emotions.

[0196] First, a user enters a search query into a device. This query is sent to a search engine on the Internet, and traditional search results are obtained. In parallel, the server sends the same search query to an artificial intelligence model that generates it, and receives the generated results from the AI.

[0197] Next, the server compares the acquired search results with the results generated by the AI. Specifically, it matches the titles and snippets of the search results with the corresponding parts of the generated results to calculate the degree of match. This degree of match serves as an index for evaluating the reliability of each search result. Furthermore, the server scores the information accuracy of each search result based on the generated results. This quantifies how closely the AI's results match the actual search results.

[0198] Based on these relevance and information accuracy scores, the server re-ranks all search results. For example, a search result with a relevance score of 90 and an information accuracy score of 95 will receive a higher overall score of 92.5. In this way, the re-ranked search results are determined.

[0199] Furthermore, this system incorporates an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's emotions through voice input and facial recognition data when the user enters a search query. This emotion information is used to filter search results and adjust the display order. For example, if the user is angry, search results that emphasize information accuracy will be displayed at the top, providing results that will help the user solve their problem. On the other hand, if the user is relaxed, general information and interesting content will be displayed preferentially.

[0200] This allows users to quickly and accurately obtain the most appropriate information according to their emotions. As a specific example, if a user enters the query "capital of Japan," the server will evaluate "Tokyo" as having a high degree of match and information accuracy, and then adjust the display order based on the user's emotions. For example, if the user is in a hurry, it is possible to provide information quickly by displaying search results for "Tokyo" at the top.

[0201] This system combines the advanced natural language processing capabilities of the artificial intelligence model it generates with the user emotion recognition capabilities of the emotion engine to deliver more reliable search results and information that is optimized for users compared to conventional search engines.

[0202] The processing flow will be explained below.

[0203] Step 1:

[0204] The user enters a search query into the device. The user clicks the "Search" button to submit the query.

[0205] Step 2:

[0206] The terminal sends the user's search query to the server.

[0207] Step 3:

[0208] The server receives the user's search query and logs the query.

[0209] Step 4:

[0210] The server sends the received search query to the search engine, and waits for a response from the search engine.

[0211] Step 5:

[0212] The search engine returns the search results to the server, which temporarily stores the received search results in its internal database.

[0213] Step 6:

[0214] The server sends the same search query to the AI ​​model, and the server waits for the AI ​​model to generate the results.

[0215] Step 7:

[0216] The generating AI model sends the generated results back to the server, which stores the received results in its internal database.

[0217] Step 8:

[0218] The server compares the search results it has obtained with the results generated by the AI. Specifically, it matches the title and snippet of the search results with the generated results.

[0219] Step 9:

[0220] The server calculates the degree of match for each search result, which is a numerical representation of the degree of match between the content of the search result and the generated result.

[0221] Step 10:

[0222] The server assigns a score to each search result based on its accuracy, calculated on a scale of 0 to 100.

[0223] Step 11:

[0224] The server re-ranks all search results based on relevance and information accuracy, calculates new rankings, and forms a re-ranked search result list.

[0225] Step 12:

[0226] The device recognizes the emotion expressed when the user enters a query through an emotion engine, and acquires emotion data through voice input and facial recognition using a camera.

[0227] Step 13:

[0228] The emotion engine analyzes the user's emotions and sends the emotion data to the server, where it is stored in a database.

[0229] Step 14:

[0230] The server adjusts the display order and content of search results based on the emotional data. For example, if the user is feeling stressed, it will prioritize displaying information that is more important.

[0231] Step 15:

[0232] The server sends the re-ranked search result list to the terminal.

[0233] Step 16:

[0234] The terminal displays the final search results to the user, who then checks the results and obtains the information they need.

[0235] Step 17:

[0236] Users can click on search results to view details, which allows them to efficiently obtain the most reliable information that best suits their emotions.

[0237] Through these processing steps, the system can provide optimal search results that take into account the user's sentiment and search accuracy. For example, if a user enters the query "capital of Japan," the server will evaluate "Tokyo" as having a high degree of match and information accuracy, and then adjust the display order based on the user's sentiment. For example, if the user is in a hurry, the server can quickly provide information by displaying search results for "Tokyo" at the top.

[0238] Example 2

[0239] 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."

[0240] Conventional search engines have difficulty providing reliable information in response to user-entered search queries, resulting in the inability of users to quickly and accurately obtain the information they need. Furthermore, conventional systems do not display search results that take the user's emotional state into account, making improving the user experience a challenge. Furthermore, there is a demand for more advanced information provision by integrating and utilizing search results with results generated by generative AI models.

[0241] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0242] In this invention, the server includes means for receiving a search query entered by a user, means for obtaining search results from a search technology based on the search query, means for transmitting the search query to an AI model that generates the search query and obtaining the generated results, means for comparing the search results with the generated results and calculating a degree of match, means for scoring the information accuracy of the search results based on the generated results, means for re-ranking the search results based on the degree of match and the information accuracy, means for recognizing and analyzing the user's emotions, means for filtering and adjusting the re-ranked search results based on the emotions, and means for displaying the re-ranked search results on a terminal, thereby enabling a user to quickly and accurately obtain reliable search results that correspond to their emotional state.

[0243] A "search query" refers to a string or phrase that a user enters to obtain information.

[0244] "Search technology" refers to algorithms and systems for retrieving information from the Internet or within databases.

[0245] An "artificial intelligence model" refers to a program or system that performs natural language processing and pattern recognition based on large amounts of data.

[0246] "Generated results" refers to the answers or information generated by an AI model based on an input search query.

[0247] "Matchability" refers to an index that numerically represents how closely the search results match the generated results.

[0248] "Information accuracy" refers to an indicator that evaluates how accurate and reliable search results or generated results are.

[0249] "Reranking" refers to the re-ordering of search results and generated results based on specified criteria.

[0250] An "emotion engine" refers to a system that analyzes emotions based on the user's voice input and facial recognition data.

[0251] "Filtering" refers to the process of sorting and ordering information based on specific criteria.

[0252] The present invention combines a system that utilizes search technology and artificial intelligence models to generate optimized search results based on a user-entered search query with an emotion engine that recognizes the user's emotions. This system performs multiple processing steps in succession to highly optimize the user's search experience.

[0253] First, a user enters a search query into a device. For example, the user enters the query "Today's weather in Tokyo." This input is entered into the search bar and confirmed by pressing the Enter key. Next, the device sends this entered search query to a search technology on the Internet (e.g., a general search engine). At the same time, the server sends the same query to an artificial intelligence model (e.g., a generative AI specialized in natural language processing) that generates it.

[0254] The server receives both the search results returned by the search engine and the generated results provided by the generation AI. For example, the search engine may return 12 search results, and the generation AI may return a sentence-format answer. The server then compares these results by matching the titles and snippets of the search results with the corresponding parts of the generated results and calculating the degree of match. The degree of match is a numerical indicator of how closely the search results and the output of the generation AI match.

[0255] Furthermore, the server scores the information accuracy of each search result based on this degree of match. For example, if a search result has a degree of match of 90 and an information accuracy of 95, its overall score will be (90 + 95) / 2 = 92.5. This allows the server to re-rank all search results. Search results with higher information accuracy will be displayed at the top, allowing users to quickly obtain more reliable information.

[0256] The system also incorporates an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's emotions through voice input and facial recognition data when the user enters a search query. For example, if the user is angry when entering a search query, the emotion engine detects this state and filters the search results to display search results that emphasize information accuracy. In addition, if the user is relaxed, general information and interesting content are prioritized.

[0257] The search results, filtered and adjusted in this way, are finally sent from the server to the device and displayed on the screen, allowing the user to quickly and accurately obtain the most appropriate information according to their emotional state.

[0258] Specific examples

[0259] When a user searches for "events in Tokyo next weekend," the following process occurs:

[0260] 1. The user types "Events in Tokyo next weekend" into the terminal.

[0261] 2. The device sends this query to a search technology (e.g., a general search engine), and the server sends the same query to an artificial intelligence model that generates it (e.g., a generative AI specialized in natural language processing).

[0262] 3. The server receives search results from the search engine and generated results from the generation AI.

[0263] 4. The server compares the received results and scores them based on their matching and accuracy. For example, the matching score is 85 and the accuracy is 90.

[0264] 5. The server re-ranks the search results based on these scores. A search result with a relevance of 85 and information accuracy of 90 will be ranked higher with an overall score of 87.5.

[0265] 6. The device sends the user's voice input and facial recognition data to the emotion engine to determine the user's emotions.

[0266] 7. The server adjusts the search results based on the analysis results of the emotion engine. If the user is relaxed, interesting event information will be displayed at the top.

[0267] 8. Finally, the device displays the filtered search results to the user, allowing them to relax and enjoy information about events in Tokyo next weekend.

[0268] This system allows users to obtain highly reliable and accurate search results in a format that is optimal for their emotional state.

[0269] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0270] Step 1:

[0271] A user enters a search query.

[0272] The user enters the prompt phrase "Today's weather in Tokyo" into the search bar on the device. Once this input is confirmed, the device prepares to send the search query to the Internet.

[0273] Input: The search query entered by the user (e.g., "Today's weather in Tokyo")

[0274] Output: The device is ready to send a search query.

[0275] Step 2:

[0276] Submit queries to search technology and generative AI.

[0277] The device sends the entered search query to the search technology, and at the same time, the server sends the same query to an artificial intelligence model that generates it (for example, an internet search engine is used as the search technology, and a natural language processing model is used as the generation AI).

[0278] Input: The search query entered

[0279] Output: Requests sent to search techniques and generative AI models

[0280] Step 3:

[0281] Receive search results and generated AI results.

[0282] The server receives both the search results returned by the search engine and the generated results provided by the generative AI model. For example, the search engine returns 12 search results and the generative AI returns one generated result.

[0283] Input: Results from search techniques and generative AI models

[0284] Output: Both result sets (e.g. 12 search results from the search engine and 1 generated result from the generative AI)

[0285] Step 4:

[0286] The search results are compared with the results of the generated AI to calculate the degree of match.

[0287] The server compares the title and snippet of the generated result with the received search result and calculates the degree of match. The degree of match is expressed as a numerical value indicating the degree of content match between the search result and the generated result. For example, the degree of match between a specific search result and the generated result is evaluated as 90.

[0288] Input: Search results, generated results

[0289] Output: Matching score (e.g. 90)

[0290] Step 5:

[0291] Scoring the accuracy of information in search results.

[0292] The server assigns a score to each search result based on its relevance. For example, if a search result has a relevance score of 90 and an information accuracy score of 95, its overall score will be (90 + 95) / 2 = 92.5.

[0293] Input: Search results, match rate

[0294] Output: Information accuracy score (e.g. 92.5)

[0295] Step 6:

[0296] Re-rank search results.

[0297] The server re-ranks all search results based on relevance and accuracy. Search results with higher accuracy are ranked higher. For example, a search result with a relevance of 85 and accuracy of 90 would be ranked higher with an overall score of 87.5.

[0298] Input: Information accuracy score

[0299] Output: Re-ranked search results

[0300] Step 7:

[0301] Analyze user sentiment.

[0302] The device sends voice input and facial recognition data when the user enters a search query to the emotion engine, which then analyzes the user's emotions based on this data. For example, it can determine that the user is angry based on their voice tone and facial expression.

[0303] Input: Voice data, facial recognition data

[0304] Output: User's emotional state (e.g. angry)

[0305] Step 8:

[0306] Filter and tailor search results based on sentiment.

[0307] The server reevaluates the search results based on the analysis results of the emotion engine. For example, if the user is angry, the server will filter the search results so that more accurate information is displayed at the top. On the other hand, if the user is relaxed, more interesting information will be displayed at the top.

[0308] Input: Re-ranked search results, user emotional state

[0309] Output: Filtered and refined search results

[0310] Step 9:

[0311] The final result is displayed to the user.

[0312] The server then sends the filtered and adjusted search results to the device, which then displays them on the screen, allowing users to quickly and accurately obtain the most appropriate information based on their emotional state.

[0313] Input: Filtered and refined search results

[0314] Output: Final search results displayed on your device

[0315] As described above, the system performs multiple processing steps in succession to highly optimize the user's search experience.

[0316] (Application example 2)

[0317] 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."

[0318] Modern search engines provide highly accurate search results based on user search queries, but they often fall short of providing optimal information tailored to the user's emotions and circumstances. In particular, online shopping sites require technologies that can efficiently meet user needs, as users' needs vary in terms of the products and information they seek. Conventional search systems are unable to re-rank or filter search results based on the user's emotions and circumstances while ensuring the reliability and accuracy of information, making it difficult to improve the user experience. To address these issues, an improved search system using an emotion engine is needed.

[0319] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0320] In this invention, the server includes means for receiving a search query, means for obtaining search results from a search engine based on the search query, means for transmitting the search query to an artificial intelligence model that generates the search query and obtaining the generated results, means for comparing the search results with the generated results and calculating the degree of match, means for scoring the information accuracy of the search results based on the generated results, means for re-ranking the search results based on the degree of match and the information accuracy, means for displaying the re-ranked search results on a terminal, means for analyzing a user's emotions, and means for filtering the search results or adjusting the display order based on the emotion information. This makes it possible to display optimal search results according to the user's emotions and situation while ensuring the reliability and information accuracy of the search results.

[0321] A "search query" is a set of character strings or phrases that a user enters when searching for information.

[0322] A "search engine" is a system that collects information on the Internet and provides the most appropriate search results in response to a search query.

[0323] A "generative artificial intelligence model" is an algorithm or system that generates text or data using deep learning or natural language processing techniques.

[0324] The "degree of match" is an index that numerically represents how closely the search results match the generated results.

[0325] "Information accuracy" is a measure for evaluating how accurate search results are with respect to the information a user is looking for.

[0326] "Reranking" is the process of reevaluating search results based on specific criteria and updating the display order.

[0327] A "terminal" is a device, such as a smartphone or computer, that a user uses to input information and receive results.

[0328] "Means for analyzing emotions" refers to technologies and systems that identify and analyze a user's emotional state from their input, facial expressions, voice, etc.

[0329] "Filtering" is the process of sorting search results based on specific criteria and removing unnecessary information.

[0330] MODE FOR CARRYING OUT THE INVENTION

[0331] To implement this invention, a system is required in which a server and a user cooperate through a terminal and provide optimal search results based on the user's search query. This system includes the following technologies and processes:

[0332] Server Processing

[0333] The server first receives a search query entered by a user on a device. This query is then sent as a request to a search engine, which retrieves traditional search results. At the same time, the server also sends the same search query to an artificial intelligence model (e.g., GPT-3) that generates it, and receives the generated results.

[0334] Next, the server compares the retrieved search results with the generated results and calculates the degree of match. Specifically, it verifies the title and snippet of the search results with the corresponding parts of the generated results and quantifies the degree of match. This degree of match serves as an index for evaluating the reliability of each search result.

[0335] Furthermore, the server scores each search result based on its generated results, quantifying how closely the generated AI results match the actual search results. Based on these match and accuracy scores, the server re-ranks all search results.

[0336] User sentiment analysis

[0337] When a user enters a search query, the device sends voice input and facial recognition data to the server. The server then uses an emotion engine to analyze the user's emotions. This emotion information is used to filter search results and adjust the display order. For example, if the user is feeling stressed, search results that emphasize information accuracy will be displayed at the top, providing results that will help the user solve their problem. On the other hand, if the user is relaxed, general information and interesting content will be displayed preferentially.

[0338] Specific processing examples

[0339] For example, if a user enters the search query "cheap laptops," the server receives search results from the search engine and results generated by the generative AI model. Then, these results are compared, scored for consistency and information accuracy, and re-ranked. Finally, if the emotion engine determines that the user's emotion is "hurry," the server prioritizes the display of the most highly rated products.

[0340] Examples of prompts that can be used include the following:

[0341] User search query: "cheap laptop"

[0342] Emotion: "I'm in a hurry"

[0343] Search result title: "cheapest laptops 2023"

[0344] Generated AI result snippet: "The cheapest laptops in 2023 are all-in-one..."

[0345] Hardware and software used

[0346] The following hardware and software are used to implement this invention:

[0347] Server: A high-performance computer for data processing and result generation

[0348] Emotion Engine: Software for Analyzing User Emotions

[0349] Generative AI models: Algorithms for natural language processing (e.g., GPT-3)

[0350] Device: A smartphone or computer where users input information and receive results.

[0351] This allows users to quickly and accurately obtain optimal search results based on their own emotions.

[0352] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0353] Step 1:

[0354] A user enters a search query into a device.

[0355] Input: User's search query (e.g. "cheap laptops")

[0356] Output: Search query sent from device to server

[0357] How it works: The user enters the information they want to find into the device's search bar and presses the search button. The device then sends this search query to the server.

[0358] Step 2:

[0359] A server receives a search query and sends it to a search engine to obtain conventional search results.

[0360] Input: Search query

[0361] Output: Search results from a search engine

[0362] Specific operation: The server sends the received search query to the search engine and retrieves search results including relevant indexes and links.

[0363] Step 3:

[0364] The server sends the same search query to the generative AI model and receives the generated results.

[0365] Input: Search query

[0366] Output: The generated results from the generative AI model (e.g., text generated by GPT-3)

[0367] How it works: The server sends search queries to the generative AI model and receives the generated text and data from the model.

[0368] Step 4:

[0369] The server compares the search results it has acquired with the generated results and calculates the degree of match.

[0370] Input: Search results, generated results

[0371] Output: Match score (e.g. title or snippet match)

[0372] What it does: The server matches the title and snippet of the search result with the generated result and calculates a score based on how well it matches.

[0373] Step 5:

[0374] The server scores the accuracy of the information in the search results based on the generated results.

[0375] Input: Search results, generated results

[0376] Output: Information accuracy score

[0377] How it works: The server evaluates how accurate the search results are based on the generated results and assigns an information accuracy score to each result.

[0378] Step 6:

[0379] The server re-ranks the search results based on match and information accuracy.

[0380] Input: Match score, information accuracy score

[0381] Output: Re-ranked search results

[0382] Specific operation: The server re-evaluates the search results based on the degree of match and information accuracy, and ranks them in a new order.

[0383] Step 7:

[0384] The device sends voice input and facial recognition data to a server to analyze the user's emotions.

[0385] Input: User voice data, facial recognition data

[0386] Output: Parsed emotion data

[0387] How it works: The device captures the user's voice and face with a camera and sends the data to a server, where the emotion engine analyzes it to determine the user's emotional state.

[0388] Step 8:

[0389] The server filters or adjusts the display order of search results based on the user's emotional information.

[0390] Input: Reranked search results, sentiment data

[0391] Output: Final search results according to sentiment

[0392] Specific operation: Based on the emotional information analyzed by the emotion engine, the server further filters the search results and displays them in the order that best suits the user's current emotions.

[0393] Step 9:

[0394] The server displays the final search results on the device.

[0395] Input: Final search results according to sentiment

[0396] Output: The final search results displayed on your terminal

[0397] Specific operation: The server sends the filtered and re-ranked search results to the terminal and displays them to the user.

[0398] Examples of prompts:

[0399] User search query: "cheap laptop"

[0400] Emotion: "I'm in a hurry"

[0401] Search result title: "cheapest laptops 2023"

[0402] Generated AI result snippet: "The cheapest laptops in 2023 are all-in-one..."

[0403] 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.

[0404] 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.

[0405] 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.

[0406] [Second embodiment]

[0407] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0408] 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.

[0409] 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).

[0410] 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.

[0411] 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.

[0412] 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).

[0413] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0414] 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.

[0415] 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.

[0416] 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.

[0417] 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.

[0418] 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."

[0419] The present invention is a system for optimizing search results based on a search query entered by a user using a search engine and a generating artificial intelligence model. The system performs the following consecutive processes: receiving a search query, obtaining search results, obtaining generated results from the generating AI, comparing and scoring the results, re-ranking, and displaying the final results.

[0420] First, a user enters a search query into a device. This query is sent to a search engine on the Internet, and traditional search results are obtained. In parallel, the server sends the same search query to an artificial intelligence model that generates it, and receives the generated results from the AI.

[0421] Next, the server compares the acquired search results with the results generated by the AI. Specifically, it matches the titles and snippets of the search results with the corresponding parts of the generated results to calculate the degree of match. This degree of match serves as an index for evaluating the reliability of each search result. Furthermore, the server scores the information accuracy of each search result based on the generated results. This quantifies how closely the AI's results match the actual search results.

[0422] Based on these relevance and information accuracy scores, the server re-ranks all search results. For example, a search result with a relevance score of 90 and an information accuracy score of 95 will receive a higher overall score of 92.5. In this way, the re-ranked search results are determined.

[0423] Finally, the server sends the re-ranked search results to the device and displays them to the user. This allows users to quickly obtain information with high reliability and guaranteed accuracy. For example, if a user enters the query "capital of Japan," the server will evaluate "Tokyo" as having a high degree of match and information accuracy and display it at the top.

[0424] By utilizing the advanced natural language processing capabilities of the AI ​​model it generates, this system is able to provide more reliable search results than conventional search engines, allowing users to quickly and accurately obtain the information they need.

[0425] The processing flow will be explained below.

[0426] Step 1:

[0427] The user enters a search query into the device. The user clicks the "Search" button to submit the query.

[0428] Step 2:

[0429] The server receives the user's search query and logs the query.

[0430] Step 3:

[0431] The server sends the received search query to the search engine, and waits for a response from the search engine.

[0432] Step 4:

[0433] The search engine returns the search results to the server, which temporarily stores the received search results in its internal database.

[0434] Step 5:

[0435] The server sends the same search query to the AI ​​model, and the server waits for the AI ​​model to generate the results.

[0436] Step 6:

[0437] The generating AI model sends the generated results back to the server, which stores the received results in its internal database.

[0438] Step 7:

[0439] The search results obtained by the server are compared with the results generated by the AI. The comparison mainly focuses on the titles and snippets of the search results.

[0440] Step 8:

[0441] The server calculates a match score for each search result, which involves quantifying the match between the content of the search result and the generated result.

[0442] Step 9:

[0443] The server scores each search result based on its accuracy, which is based on how reliable the generated result is.

[0444] Step 10:

[0445] The server re-ranks all search results based on relevance and information accuracy, calculates new rankings, and forms a re-ranked search result list.

[0446] Step 11:

[0447] The server sends the re-ranked search result list to the terminal, and the terminal displays the received search results for the user to review.

[0448] Step 12:

[0449] The user checks the displayed search results and obtains the information they need. The user clicks on the search results to view the details.

[0450] Example 1

[0451] 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."

[0452] Modern information retrieval systems often provide irrelevant or unreliable information in response to user-entered search queries. This prevents users from quickly obtaining the information they need, reducing search efficiency. Furthermore, traditional information retrieval services lack the means to adequately evaluate the reliability of search results, exposing users to the risk of making decisions based on incorrect information.

[0453] 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.

[0454] In this invention, the server includes means for receiving a search query, means for obtaining search results from an information search service, and means for sending a query to a generating artificial intelligence model and obtaining the generated results, thereby making it possible to calculate the degree of match between the obtained search results and the generated results, score the information accuracy based on that, and provide the user with re-ranked search results.

[0455] A "search query" is a string of characters that a user enters to search for information.

[0456] An "information search service" is a system or engine for obtaining information on the Internet, generally referring to a search engine.

[0457] A "generative artificial intelligence model" is a machine learning model for generating text based on a given query.

[0458] The "generation result" is the text data that the generating artificial intelligence model outputs in response to the input query.

[0459] "Search results" refers to the relevant information returned by an information retrieval service in response to a search query.

[0460] The "degree of match" is a numerical value or index that evaluates the degree of match between the search result and the generated result.

[0461] "Information accuracy" is a numerical value or index that indicates how accurate and reliable the generated results are.

[0462] "Reranking" is the process of reevaluating and rearranging the ranking of search results based on relevance and information accuracy.

[0463] "Terminal" refers to the device on which a user enters a search query and displays results, including a PC or smartphone.

[0464] This invention is a system for optimizing search results based on a user-provided search query by utilizing an information search service and a generative AI model. The system performs the following consecutive processes: receiving a search query, obtaining search results, obtaining generated results from a generative AI model, comparing the results and calculating the degree of match, scoring the information accuracy, re-ranking, and displaying the final results.

[0465] First, a user inputs a search query into a terminal. For example, the user inputs the query "capital of Japan." This query is sent to an information search service (e.g., a general-purpose search engine) via the Internet. At the same time, the same query is sent to a server.

[0466] The server sends the received search query to a generative AI model (e.g., a natural language processing model). Specifically, OpenAI's GPT-3 is used. The generative AI model generates text based on the query and returns the generated results to the server.

[0467] Next, the server compares the search results obtained from the information search service with the results generated by the AI. This comparison is performed by matching the titles and summaries of the search results with the corresponding parts of the generated results. The server then calculates the degree of match. The degree of match is a numerical value that evaluates the degree of match between the search results and the generated results.

[0468] Next, the server performs a scoring process. Based on the results generated by the AI, the server scores the accuracy of each search result. Information accuracy is a numerical value that indicates how accurate and reliable the generated results are.

[0469] Based on these relevance and accuracy scores, the server re-ranks all search results. For example, if a search result has a relevance score of 90 and an accuracy score of 95, the overall score will be calculated as 92.5. Search results with higher scores will ultimately be ranked higher.

[0470] Finally, the server sends the re-ranked search results to the terminal, which then displays the re-ranked search results to the user, allowing the user to quickly check highly reliable and accurate information.

[0471] For example, if a user enters the query "capital of Japan," the server compares the results obtained from the information search service and the generative AI model, and scores search results containing "Tokyo" with high matching and information accuracy. As a result of the re-ranking process, information related to "Tokyo" is displayed at the top.

[0472] An example prompt is:

[0473] Query: capital of japan

[0474] Explanation: What is the capital of Japan?

[0475] By implementing this invention, users can quickly and accurately obtain the information they need. By utilizing the advanced natural language processing capabilities of the generated artificial intelligence model, it is possible to provide more reliable search results than conventional information search services.

[0476] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0477] Step 1:

[0478] User enters a search query

[0479] A user enters a search query into a device. For example, the user enters "capital of Japan."

[0480] Input: Search query (e.g. "Capital of Japan")

[0481] Output: Search queries sent from the device to the internet

[0482] Step 2:

[0483] Device submitting search queries

[0484] The terminal transmits the search query entered by the user to the information search service via the Internet, and also transmits the same query to the server.

[0485] Input: User's search query

[0486] Output: Search query sent to information retrieval service and server

[0487] Step 3:

[0488] Server obtains search results from information search service

[0489] The server retrieves search results from the information search service based on the submitted search query, using an API to retrieve results from a search engine (e.g., a general-purpose search engine).

[0490] Input: Search results from information search services

[0491] Output: Search results saved on the server

[0492] Step 4:

[0493] Server sends queries to the generative AI model

[0494] The server sends the received search query to a generative AI model, such as a natural language processing model (e.g., OpenAI's GPT-3).

[0495] Input: Search query

[0496] Output: Request to the generative AI model

[0497] Step 5:

[0498] Obtaining generated results from the generative AI model by the server

[0499] The generative AI model generates text based on the search query and returns the results to the server, which receives the generated results.

[0500] Input: Response from a generative AI model

[0501] Output: Generated results stored on the server

[0502] Step 6:

[0503] Comparing server search results with generated results

[0504] The server compares the search results from the information retrieval service with the generated results from the generative AI model by matching the titles and summaries of the search results with the corresponding parts of the generated results.

[0505] Input: Search results and generated results

[0506] Data processing: Matching with text comparison algorithms

[0507] Output: Match score for each result

[0508] Step 7:

[0509] Server-based scoring of information accuracy

[0510] The server scores the information accuracy of each search result based on the results generated by the AI. Information accuracy is a numerical value that evaluates the accuracy and reliability of the generated results.

[0511] Input: Generated results

[0512] Data Computation: Information Accuracy Scoring Algorithm

[0513] Output: Information accuracy score for each search result

[0514] Step 8:

[0515] Server Re-ranking

[0516] The server re-ranks all search results based on their relevance and accuracy scores, with more reliable results being ranked higher.

[0517] Input: Match score and information accuracy score

[0518] Data arithmetic: Re-ranking by weighted average of scores

[0519] Output: Re-ranked search results

[0520] Step 9:

[0521] Server sends re-ranked search results

[0522] The server transmits the re-ranked search results to the terminal.

[0523] Input: Reranked search results

[0524] Output: Search results sent to your device

[0525] Step 10:

[0526] Displaying search results by device

[0527] The terminal displays the re-ranked search results to the user, who can then check the displayed search results and quickly and accurately obtain the information they need.

[0528] Input: Reranked search results

[0529] Output: Search results displayed to the user

[0530] (Application example 1)

[0531] 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."

[0532] In conventional food delivery services, even when a user inputs the name of a desired dish or restaurant, the accuracy and reliability of the search results are insufficient, making it difficult to quickly find the best option. Furthermore, the search results often contain advertisements and irrelevant information, hindering the user's ability to select the best restaurant. The objective of the present invention is to solve these problems and quickly and accurately provide the information desired by the user.

[0533] 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.

[0534] In this invention, the server includes means for receiving a search query, means for obtaining search results from a search engine based on the search query, means for transmitting the search query to an AI model that generates the search query and obtaining the generated results, means for comparing the search results with the generated results and calculating a degree of match, means for scoring the information accuracy of the search results based on the generated results, means for re-ranking the search results based on the degree of match and the information accuracy, means for displaying the re-ranked search results on a terminal, and means for presenting food delivery options optimized using the AI ​​model based on the name of a dish or restaurant entered by a user. This enables a user to quickly obtain more reliable and accurate food delivery information in response to a search query.

[0535] A "search query" is a keyword or phrase that a user enters when searching for information.

[0536] A "search engine" is a system that searches for information on the Internet and provides related web pages and data.

[0537] A "generative artificial intelligence model" is an AI algorithm that performs natural language processing based on user input and generates information.

[0538] The "degree of match" is an index for evaluating the degree of match between the search results and the generated results.

[0539] "Information accuracy" is a score that indicates the accuracy and reliability of the information calculated by the generative AI model.

[0540] "Re-ranking" is a process of re-ranking the acquired search results based on the scores of match and information accuracy.

[0541] A "terminal" is an electronic device such as a personal computer, smartphone, or tablet.

[0542] The "dish name" is the name of a specific food item that a user specifies when using a food delivery service.

[0543] "Restaurant name" refers to the name of the restaurant or cafe that provides the food delivery service.

[0544] "Food delivery options" refers to multiple delivery service options offered for the food or restaurant searched for by the user.

[0545] The present invention provides a system for optimizing search results for food delivery services based on a search query entered by a user. The system supports a series of processes including receiving a search query, obtaining search results, obtaining generated results using an artificial intelligence model, comparing and scoring both results, re-ranking, and displaying the final results.

[0546] Hardware and Software Configuration

[0547] Server: A server with powerful processing power is responsible for processing search queries and aggregating results.

[0548] Device: A device such as a personal computer, smartphone, or tablet is used as the device where the user enters the search query and displays the final results.

[0549] software:

[0550] Programming language: Python

[0551] Libraries: Requests library (used to communicate with search engine APIs and food delivery APIs)

[0552] Generative AI models: Leverage artificial intelligence generative models (e.g., GPT-3) through the OpenAI API.

[0553] Data processing and calculation flow

[0554] The server first receives a search query entered by a user on a device, sends the search query to a food delivery search engine, and retrieves relevant search results. In parallel, the same search query is sent to a generative artificial intelligence model, and retrieves generated results.

[0555] The server then compares the search engine results with the results generated by the generative AI model and calculates the degree of match between them. This degree of match is calculated based on the degree of match between the title and snippet of the search result and the title and snippet of the generated result. The information accuracy of the search result is also scored based on the reliability of the generated result.

[0556] The server re-ranks the search results based on the match and information accuracy scores, and the re-ranked search results are sent to the terminal and displayed to the user.

[0557] Specific examples

[0558] For example, if a user types "sushi" into their device, the server retrieves food delivery options related to "sushi" from the search engine and the generative AI model. Then, it compares both results and re-ranks them based on match and information accuracy. Finally, the user is presented with the most reliable and accurate sushi delivery options.

[0559] Prompt Sentence Examples

[0560] "Give me the best food delivery options for Sushi"

[0561] Such a system allows users to obtain reliable food delivery information more efficiently.

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

[0563] Step 1:

[0564] A user inputs a search query into a terminal. The input search query is the name of a dish or restaurant that the user is interested in, such as "sushi." The input query becomes the basis for other processing steps in the system.

[0565] Step 2:

[0566] The server receives a search query. The received search query is sent to a search engine API to retrieve relevant search results. For example, if the search query is "sushi," the server retrieves multiple food delivery options that offer sushi. The input is the search query "sushi," and the output is a list of relevant search results.

[0567] Step 3:

[0568] The server sends the same search query to a generative AI model. The generative AI model provides generated results based on the query, for example using the OpenAI API. The input is the search query "sushi," and the output is a list of recommended food delivery options provided by the generative AI. The prompt provided by the generative AI model is "Give me the best food delivery options for Sushi."

[0569] Step 4:

[0570] The server compares the search results with the results of the generative AI model. Specifically, it compares the titles and snippets of the search results with those of the generated results and calculates the degree of match. The degree of match is quantified, for example, to show how many common elements each result contains. The input is the search results and the generated results, and the output is a score indicating the degree of match between each result.

[0571] Step 5:

[0572] The server scores the information accuracy of the search results based on the generated results. In this process, an information accuracy score is assigned to each search result based on the reliability index of the generative AI model and the content of the search result. The input is the match score and the reliability index of the generative AI model, and the output is the information accuracy score for each search result.

[0573] Step 6:

[0574] The server re-ranks the search results based on the relevance and accuracy scores. This re-ranking adjusts the ranking so that results with high relevance and accuracy are displayed at the top. The input is the relevance and accuracy scores of each search result, and the output is a list of re-ranked search results.

[0575] Step 7:

[0576] The server sends the re-ranked search results to the device. The user reviews these results on the device and selects the food delivery option that best suits them. The input is a list of re-ranked search results, and the output is the optimized search results displayed on the device.

[0577] 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.

[0578] The present invention combines a system that utilizes a search engine and a generating artificial intelligence model to optimize search results based on a search query entered by a user, with an emotion engine that recognizes the user's emotions. The system performs the following continuous processes: receiving a search query, obtaining search results, obtaining generated results from the generating AI, comparing and scoring the results, re-ranking, displaying the final results, and filtering and adjusting the search results based on the user's emotions.

[0579] First, a user enters a search query into a device. This query is sent to a search engine on the Internet, and traditional search results are obtained. In parallel, the server sends the same search query to an artificial intelligence model that generates it, and receives the generated results from the AI.

[0580] Next, the server compares the acquired search results with the results generated by the AI. Specifically, it matches the titles and snippets of the search results with the corresponding parts of the generated results to calculate the degree of match. This degree of match serves as an index for evaluating the reliability of each search result. Furthermore, the server scores the information accuracy of each search result based on the generated results. This quantifies how closely the AI's results match the actual search results.

[0581] Based on these relevance and information accuracy scores, the server re-ranks all search results. For example, a search result with a relevance score of 90 and an information accuracy score of 95 will receive a higher overall score of 92.5. In this way, the re-ranked search results are determined.

[0582] Furthermore, this system incorporates an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's emotions through voice input and facial recognition data when the user enters a search query. This emotion information is used to filter search results and adjust the display order. For example, if the user is angry, search results that emphasize information accuracy will be displayed at the top, providing results that will help the user solve their problem. On the other hand, if the user is relaxed, general information and interesting content will be displayed preferentially.

[0583] This allows users to quickly and accurately obtain the most appropriate information according to their emotions. As a specific example, if a user enters the query "capital of Japan," the server will evaluate "Tokyo" as having a high degree of match and information accuracy, and then adjust the display order based on the user's emotions. For example, if the user is in a hurry, it is possible to provide information quickly by displaying search results for "Tokyo" at the top.

[0584] This system combines the advanced natural language processing capabilities of the artificial intelligence model it generates with the user emotion recognition capabilities of the emotion engine to deliver more reliable search results and information that is optimized for users compared to conventional search engines.

[0585] The processing flow will be explained below.

[0586] Step 1:

[0587] The user enters a search query into the device. The user clicks the "Search" button to submit the query.

[0588] Step 2:

[0589] The terminal sends the user's search query to the server.

[0590] Step 3:

[0591] The server receives the user's search query and logs the query.

[0592] Step 4:

[0593] The server sends the received search query to the search engine, and waits for a response from the search engine.

[0594] Step 5:

[0595] The search engine returns the search results to the server, which temporarily stores the received search results in its internal database.

[0596] Step 6:

[0597] The server sends the same search query to the AI ​​model, and the server waits for the AI ​​model to generate the results.

[0598] Step 7:

[0599] The generating AI model sends the generated results back to the server, which stores the received results in its internal database.

[0600] Step 8:

[0601] The server compares the search results it has obtained with the results generated by the AI. Specifically, it matches the title and snippet of the search results with the generated results.

[0602] Step 9:

[0603] The server calculates the degree of match for each search result, which is a numerical representation of the degree of match between the content of the search result and the generated result.

[0604] Step 10:

[0605] The server assigns a score to each search result based on its accuracy, calculated on a scale of 0 to 100.

[0606] Step 11:

[0607] The server re-ranks all search results based on relevance and information accuracy, calculates new rankings, and forms a re-ranked search result list.

[0608] Step 12:

[0609] The device recognizes the emotion expressed when the user enters a query through an emotion engine, and acquires emotion data through voice input and facial recognition using a camera.

[0610] Step 13:

[0611] The emotion engine analyzes the user's emotions and sends the emotion data to the server, where it is stored in a database.

[0612] Step 14:

[0613] The server adjusts the display order and content of search results based on the emotional data. For example, if the user is feeling stressed, it will prioritize displaying information that is more important.

[0614] Step 15:

[0615] The server sends the re-ranked search result list to the terminal.

[0616] Step 16:

[0617] The terminal displays the final search results to the user, who then checks the results and obtains the information they need.

[0618] Step 17:

[0619] Users can click on search results to view details, which allows them to efficiently obtain the most reliable information that best suits their emotions.

[0620] Through these processing steps, the system can provide optimal search results that take into account the user's sentiment and search accuracy. For example, if a user enters the query "capital of Japan," the server will evaluate "Tokyo" as having a high degree of match and information accuracy, and then adjust the display order based on the user's sentiment. For example, if the user is in a hurry, the server can quickly provide information by displaying search results for "Tokyo" at the top.

[0621] Example 2

[0622] 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."

[0623] Conventional search engines have difficulty providing reliable information in response to user-entered search queries, resulting in the inability of users to quickly and accurately obtain the information they need. Furthermore, conventional systems do not display search results that take the user's emotional state into account, making improving the user experience a challenge. Furthermore, there is a demand for more advanced information provision by integrating and utilizing search results with results generated by generative AI models.

[0624] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0625] In this invention, the server includes means for receiving a search query entered by a user, means for obtaining search results from a search technology based on the search query, means for transmitting the search query to an AI model that generates the search query and obtaining the generated results, means for comparing the search results with the generated results and calculating a degree of match, means for scoring the information accuracy of the search results based on the generated results, means for re-ranking the search results based on the degree of match and the information accuracy, means for recognizing and analyzing the user's emotions, means for filtering and adjusting the re-ranked search results based on the emotions, and means for displaying the re-ranked search results on a terminal, thereby enabling a user to quickly and accurately obtain reliable search results that correspond to their emotional state.

[0626] A "search query" refers to a string or phrase that a user enters to obtain information.

[0627] "Search technology" refers to algorithms and systems for retrieving information from the Internet or within databases.

[0628] An "artificial intelligence model" refers to a program or system that performs natural language processing and pattern recognition based on large amounts of data.

[0629] "Generated results" refers to the answers or information generated by an AI model based on an input search query.

[0630] "Matchability" refers to an index that numerically represents how closely the search results match the generated results.

[0631] "Information accuracy" refers to an indicator that evaluates how accurate and reliable search results or generated results are.

[0632] "Reranking" refers to the re-ordering of search results and generated results based on specified criteria.

[0633] An "emotion engine" refers to a system that analyzes emotions based on the user's voice input and facial recognition data.

[0634] "Filtering" refers to the process of sorting and ordering information based on specific criteria.

[0635] The present invention combines a system that utilizes search technology and artificial intelligence models to generate optimized search results based on a user-entered search query with an emotion engine that recognizes the user's emotions. This system performs multiple processing steps in succession to highly optimize the user's search experience.

[0636] First, a user enters a search query into a device. For example, the user enters the query "Today's weather in Tokyo." This input is entered into the search bar and confirmed by pressing the Enter key. Next, the device sends this entered search query to a search technology on the Internet (e.g., a general search engine). At the same time, the server sends the same query to an artificial intelligence model (e.g., a generative AI specialized in natural language processing) that generates it.

[0637] The server receives both the search results returned by the search engine and the generated results provided by the generation AI. For example, the search engine may return 12 search results, and the generation AI may return a sentence-format answer. The server then compares these results by matching the titles and snippets of the search results with the corresponding parts of the generated results and calculating the degree of match. The degree of match is a numerical indicator of how closely the search results and the output of the generation AI match.

[0638] Furthermore, the server scores the information accuracy of each search result based on this degree of match. For example, if a search result has a degree of match of 90 and an information accuracy of 95, its overall score will be (90 + 95) / 2 = 92.5. This allows the server to re-rank all search results. Search results with higher information accuracy will be displayed at the top, allowing users to quickly obtain more reliable information.

[0639] The system also incorporates an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's emotions through voice input and facial recognition data when the user enters a search query. For example, if the user is angry when entering a search query, the emotion engine detects this state and filters the search results to display search results that emphasize information accuracy. In addition, if the user is relaxed, general information and interesting content are prioritized.

[0640] The search results, filtered and adjusted in this way, are finally sent from the server to the device and displayed on the screen, allowing the user to quickly and accurately obtain the most appropriate information according to their emotional state.

[0641] Specific examples

[0642] When a user searches for "events in Tokyo next weekend," the following process occurs:

[0643] 1. The user types "Events in Tokyo next weekend" into the terminal.

[0644] 2. The device sends this query to a search technology (e.g., a general search engine), and the server sends the same query to an artificial intelligence model that generates it (e.g., a generative AI specialized in natural language processing).

[0645] 3. The server receives search results from the search engine and generated results from the generation AI.

[0646] 4. The server compares the received results and scores them based on their matching and accuracy. For example, the matching score is 85 and the accuracy is 90.

[0647] 5. The server re-ranks the search results based on these scores. A search result with a relevance of 85 and information accuracy of 90 will be ranked higher with an overall score of 87.5.

[0648] 6. The device sends the user's voice input and facial recognition data to the emotion engine to determine the user's emotions.

[0649] 7. The server adjusts the search results based on the analysis results of the emotion engine. If the user is relaxed, interesting event information will be displayed at the top.

[0650] 8. Finally, the device displays the filtered search results to the user, allowing them to relax and enjoy information about events in Tokyo next weekend.

[0651] This system allows users to obtain highly reliable and accurate search results in a format that is optimal for their emotional state.

[0652] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0653] Step 1:

[0654] A user enters a search query.

[0655] The user enters the prompt phrase "Today's weather in Tokyo" into the search bar on the device. Once this input is confirmed, the device prepares to send the search query to the Internet.

[0656] Input: The search query entered by the user (e.g., "Today's weather in Tokyo")

[0657] Output: The device is ready to send a search query.

[0658] Step 2:

[0659] Submit queries to search technology and generative AI.

[0660] The device sends the entered search query to the search technology, and at the same time, the server sends the same query to an artificial intelligence model that generates it (for example, an internet search engine is used as the search technology, and a natural language processing model is used as the generation AI).

[0661] Input: The search query entered

[0662] Output: Requests sent to search techniques and generative AI models

[0663] Step 3:

[0664] Receive search results and generated AI results.

[0665] The server receives both the search results returned by the search engine and the generated results provided by the generative AI model. For example, the search engine returns 12 search results and the generative AI returns one generated result.

[0666] Input: Results from search techniques and generative AI models

[0667] Output: Both result sets (e.g. 12 search results from the search engine and 1 generated result from the generative AI)

[0668] Step 4:

[0669] The search results are compared with the results of the generated AI to calculate the degree of match.

[0670] The server compares the title and snippet of the generated result with the received search result and calculates the degree of match. The degree of match is expressed as a numerical value indicating the degree of content match between the search result and the generated result. For example, the degree of match between a specific search result and the generated result is evaluated as 90.

[0671] Input: Search results, generated results

[0672] Output: Matching score (e.g. 90)

[0673] Step 5:

[0674] Scoring the accuracy of information in search results.

[0675] The server assigns a score to each search result based on its relevance. For example, if a search result has a relevance score of 90 and an information accuracy score of 95, its overall score will be (90 + 95) / 2 = 92.5.

[0676] Input: Search results, match rate

[0677] Output: Information accuracy score (e.g. 92.5)

[0678] Step 6:

[0679] Re-rank search results.

[0680] The server re-ranks all search results based on relevance and accuracy. Search results with higher accuracy are ranked higher. For example, a search result with a relevance of 85 and accuracy of 90 would be ranked higher with an overall score of 87.5.

[0681] Input: Information accuracy score

[0682] Output: Re-ranked search results

[0683] Step 7:

[0684] Analyze user sentiment.

[0685] The device sends voice input and facial recognition data when the user enters a search query to the emotion engine, which then analyzes the user's emotions based on this data. For example, it can determine that the user is angry based on their voice tone and facial expression.

[0686] Input: Voice data, facial recognition data

[0687] Output: User's emotional state (e.g. angry)

[0688] Step 8:

[0689] Filter and tailor search results based on sentiment.

[0690] The server reevaluates the search results based on the analysis results of the emotion engine. For example, if the user is angry, the server will filter the search results so that more accurate information is displayed at the top. On the other hand, if the user is relaxed, more interesting information will be displayed at the top.

[0691] Input: Re-ranked search results, user emotional state

[0692] Output: Filtered and refined search results

[0693] Step 9:

[0694] The final result is displayed to the user.

[0695] The server then sends the filtered and adjusted search results to the device, which then displays them on the screen, allowing users to quickly and accurately obtain the most appropriate information based on their emotional state.

[0696] Input: Filtered and refined search results

[0697] Output: Final search results displayed on your device

[0698] As described above, the system performs multiple processing steps in succession to highly optimize the user's search experience.

[0699] (Application example 2)

[0700] 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."

[0701] Modern search engines provide highly accurate search results based on user search queries, but they often fall short of providing optimal information tailored to the user's emotions and circumstances. In particular, online shopping sites require technologies that can efficiently meet user needs, as users' needs vary in terms of the products and information they seek. Conventional search systems are unable to re-rank or filter search results based on the user's emotions and circumstances while ensuring the reliability and accuracy of information, making it difficult to improve the user experience. To address these issues, an improved search system using an emotion engine is needed.

[0702] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0703] In this invention, the server includes means for receiving a search query, means for obtaining search results from a search engine based on the search query, means for transmitting the search query to an artificial intelligence model that generates the search query and obtaining the generated results, means for comparing the search results with the generated results and calculating the degree of match, means for scoring the information accuracy of the search results based on the generated results, means for re-ranking the search results based on the degree of match and the information accuracy, means for displaying the re-ranked search results on a terminal, means for analyzing a user's emotions, and means for filtering the search results or adjusting the display order based on the emotion information. This makes it possible to display optimal search results according to the user's emotions and situation while ensuring the reliability and information accuracy of the search results.

[0704] A "search query" is a set of character strings or phrases that a user enters when searching for information.

[0705] A "search engine" is a system that collects information on the Internet and provides the most appropriate search results in response to a search query.

[0706] A "generative artificial intelligence model" is an algorithm or system that generates text or data using deep learning or natural language processing techniques.

[0707] The "degree of match" is an index that numerically represents how closely the search results match the generated results.

[0708] "Information accuracy" is a measure for evaluating how accurate search results are with respect to the information a user is looking for.

[0709] "Reranking" is the process of reevaluating search results based on specific criteria and updating the display order.

[0710] A "terminal" is a device, such as a smartphone or computer, that a user uses to input information and receive results.

[0711] "Means for analyzing emotions" refers to technologies and systems that identify and analyze a user's emotional state from their input, facial expressions, voice, etc.

[0712] "Filtering" is the process of sorting search results based on specific criteria and removing unnecessary information.

[0713] MODE FOR CARRYING OUT THE INVENTION

[0714] To implement this invention, a system is required in which a server and a user cooperate through a terminal and provide optimal search results based on the user's search query. This system includes the following technologies and processes:

[0715] Server Processing

[0716] The server first receives a search query entered by a user on a device. This query is then sent as a request to a search engine, which retrieves traditional search results. At the same time, the server also sends the same search query to an artificial intelligence model (e.g., GPT-3) that generates it, and receives the generated results.

[0717] Next, the server compares the retrieved search results with the generated results and calculates the degree of match. Specifically, it verifies the title and snippet of the search results with the corresponding parts of the generated results and quantifies the degree of match. This degree of match serves as an index for evaluating the reliability of each search result.

[0718] Furthermore, the server scores each search result based on its generated results, quantifying how closely the generated AI results match the actual search results. Based on these match and accuracy scores, the server re-ranks all search results.

[0719] User sentiment analysis

[0720] When a user enters a search query, the device sends voice input and facial recognition data to the server. The server then uses an emotion engine to analyze the user's emotions. This emotion information is used to filter search results and adjust the display order. For example, if the user is feeling stressed, search results that emphasize information accuracy will be displayed at the top, providing results that will help the user solve their problem. On the other hand, if the user is relaxed, general information and interesting content will be displayed preferentially.

[0721] Specific processing examples

[0722] For example, if a user enters the search query "cheap laptops," the server receives search results from the search engine and results generated by the generative AI model. Then, these results are compared, scored for consistency and information accuracy, and re-ranked. Finally, if the emotion engine determines that the user's emotion is "hurry," the server prioritizes the display of the most highly rated products.

[0723] Examples of prompts that can be used include the following:

[0724] User search query: "cheap laptop"

[0725] Emotion: "I'm in a hurry"

[0726] Search result title: "cheapest laptops 2023"

[0727] Generated AI result snippet: "The cheapest laptops in 2023 are all-in-one..."

[0728] Hardware and software used

[0729] The following hardware and software are used to implement this invention:

[0730] Server: A high-performance computer for data processing and result generation

[0731] Emotion Engine: Software for Analyzing User Emotions

[0732] Generative AI models: Algorithms for natural language processing (e.g., GPT-3)

[0733] Device: A smartphone or computer where users input information and receive results.

[0734] This allows users to quickly and accurately obtain optimal search results based on their own emotions.

[0735] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0736] Step 1:

[0737] A user enters a search query into a device.

[0738] Input: User's search query (e.g. "cheap laptops")

[0739] Output: Search query sent from device to server

[0740] How it works: The user enters the information they want to find into the device's search bar and presses the search button. The device then sends this search query to the server.

[0741] Step 2:

[0742] A server receives a search query and sends it to a search engine to obtain conventional search results.

[0743] Input: Search query

[0744] Output: Search results from a search engine

[0745] Specific operation: The server sends the received search query to the search engine and retrieves search results including relevant indexes and links.

[0746] Step 3:

[0747] The server sends the same search query to the generative AI model and receives the generated results.

[0748] Input: Search query

[0749] Output: The generated results from the generative AI model (e.g., text generated by GPT-3)

[0750] How it works: The server sends search queries to the generative AI model and receives the generated text and data from the model.

[0751] Step 4:

[0752] The server compares the search results it has acquired with the generated results and calculates the degree of match.

[0753] Input: Search results, generated results

[0754] Output: Match score (e.g. title or snippet match)

[0755] What it does: The server matches the title and snippet of the search result with the generated result and calculates a score based on how well it matches.

[0756] Step 5:

[0757] The server scores the accuracy of the information in the search results based on the generated results.

[0758] Input: Search results, generated results

[0759] Output: Information accuracy score

[0760] How it works: The server evaluates how accurate the search results are based on the generated results and assigns an information accuracy score to each result.

[0761] Step 6:

[0762] The server re-ranks the search results based on match and information accuracy.

[0763] Input: Match score, information accuracy score

[0764] Output: Re-ranked search results

[0765] Specific operation: The server re-evaluates the search results based on the degree of match and information accuracy, and ranks them in a new order.

[0766] Step 7:

[0767] The device sends voice input and facial recognition data to a server to analyze the user's emotions.

[0768] Input: User voice data, facial recognition data

[0769] Output: Parsed emotion data

[0770] How it works: The device captures the user's voice and face with a camera and sends the data to a server, where the emotion engine analyzes it to determine the user's emotional state.

[0771] Step 8:

[0772] The server filters or adjusts the display order of search results based on the user's emotional information.

[0773] Input: Reranked search results, sentiment data

[0774] Output: Final search results according to sentiment

[0775] Specific operation: Based on the emotional information analyzed by the emotion engine, the server further filters the search results and displays them in the order that best suits the user's current emotions.

[0776] Step 9:

[0777] The server displays the final search results on the device.

[0778] Input: Final search results according to sentiment

[0779] Output: The final search results displayed on your terminal

[0780] Specific operation: The server sends the filtered and re-ranked search results to the terminal and displays them to the user.

[0781] Examples of prompts:

[0782] User search query: "cheap laptop"

[0783] Emotion: "I'm in a hurry"

[0784] Search result title: "cheapest laptops 2023"

[0785] Generated AI result snippet: "The cheapest laptops in 2023 are all-in-one..."

[0786] 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.

[0787] 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.

[0788] 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.

[0789] [Third embodiment]

[0790] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0791] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0792] 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).

[0793] 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.

[0794] 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.

[0795] 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).

[0796] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0797] 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.

[0798] 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.

[0799] 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.

[0800] 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.

[0801] 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."

[0802] The present invention is a system for optimizing search results based on a search query entered by a user using a search engine and a generating artificial intelligence model. The system performs the following consecutive processes: receiving a search query, obtaining search results, obtaining generated results from the generating AI, comparing and scoring the results, re-ranking, and displaying the final results.

[0803] First, a user enters a search query into a device. This query is sent to a search engine on the Internet, and traditional search results are obtained. In parallel, the server sends the same search query to an artificial intelligence model that generates it, and receives the generated results from the AI.

[0804] Next, the server compares the acquired search results with the results generated by the AI. Specifically, it matches the titles and snippets of the search results with the corresponding parts of the generated results to calculate the degree of match. This degree of match serves as an index for evaluating the reliability of each search result. Furthermore, the server scores the information accuracy of each search result based on the generated results. This quantifies how closely the AI's results match the actual search results.

[0805] Based on these relevance and information accuracy scores, the server re-ranks all search results. For example, a search result with a relevance score of 90 and an information accuracy score of 95 will receive a higher overall score of 92.5. In this way, the re-ranked search results are determined.

[0806] Finally, the server sends the re-ranked search results to the device and displays them to the user. This allows users to quickly obtain information with high reliability and guaranteed accuracy. For example, if a user enters the query "capital of Japan," the server will evaluate "Tokyo" as having a high degree of match and information accuracy and display it at the top.

[0807] By utilizing the advanced natural language processing capabilities of the AI ​​model it generates, this system is able to provide more reliable search results than conventional search engines, allowing users to quickly and accurately obtain the information they need.

[0808] The processing flow will be explained below.

[0809] Step 1:

[0810] The user enters a search query into the device. The user clicks the "Search" button to submit the query.

[0811] Step 2:

[0812] The server receives the user's search query and logs the query.

[0813] Step 3:

[0814] The server sends the received search query to the search engine, and waits for a response from the search engine.

[0815] Step 4:

[0816] The search engine returns the search results to the server, which temporarily stores the received search results in its internal database.

[0817] Step 5:

[0818] The server sends the same search query to the AI ​​model, and the server waits for the AI ​​model to generate the results.

[0819] Step 6:

[0820] The generating AI model sends the generated results back to the server, which stores the received results in its internal database.

[0821] Step 7:

[0822] The search results obtained by the server are compared with the results generated by the AI. The comparison mainly focuses on the titles and snippets of the search results.

[0823] Step 8:

[0824] The server calculates a match score for each search result, which involves quantifying the match between the content of the search result and the generated result.

[0825] Step 9:

[0826] The server scores each search result based on its accuracy, which is based on how reliable the generated result is.

[0827] Step 10:

[0828] The server re-ranks all search results based on relevance and information accuracy, calculates new rankings, and forms a re-ranked search result list.

[0829] Step 11:

[0830] The server sends the re-ranked search result list to the terminal, and the terminal displays the received search results for the user to review.

[0831] Step 12:

[0832] The user checks the displayed search results and obtains the information they need. The user clicks on the search results to view the details.

[0833] Example 1

[0834] 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."

[0835] Modern information retrieval systems often provide irrelevant or unreliable information in response to user-entered search queries. This prevents users from quickly obtaining the information they need, reducing search efficiency. Furthermore, traditional information retrieval services lack the means to adequately evaluate the reliability of search results, exposing users to the risk of making decisions based on incorrect information.

[0836] 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.

[0837] In this invention, the server includes means for receiving a search query, means for obtaining search results from an information search service, and means for sending a query to a generating artificial intelligence model and obtaining the generated results, thereby making it possible to calculate the degree of match between the obtained search results and the generated results, score the information accuracy based on that, and provide the user with re-ranked search results.

[0838] A "search query" is a string of characters that a user enters to search for information.

[0839] An "information search service" is a system or engine for obtaining information on the Internet, generally referring to a search engine.

[0840] A "generative artificial intelligence model" is a machine learning model for generating text based on a given query.

[0841] The "generation result" is the text data that the generating artificial intelligence model outputs in response to the input query.

[0842] "Search results" refers to the relevant information returned by an information retrieval service in response to a search query.

[0843] The "degree of match" is a numerical value or index that evaluates the degree of match between the search result and the generated result.

[0844] "Information accuracy" is a numerical value or index that indicates how accurate and reliable the generated results are.

[0845] "Reranking" is the process of reevaluating and rearranging the ranking of search results based on relevance and information accuracy.

[0846] "Terminal" refers to the device on which a user enters a search query and displays results, including a PC or smartphone.

[0847] This invention is a system for optimizing search results based on a user-provided search query by utilizing an information search service and a generative AI model. The system performs the following consecutive processes: receiving a search query, obtaining search results, obtaining generated results from a generative AI model, comparing the results and calculating the degree of match, scoring the information accuracy, re-ranking, and displaying the final results.

[0848] First, a user inputs a search query into a terminal. For example, the user inputs the query "capital of Japan." This query is sent to an information search service (e.g., a general-purpose search engine) via the Internet. At the same time, the same query is sent to a server.

[0849] The server sends the received search query to a generative AI model (e.g., a natural language processing model). Specifically, OpenAI's GPT-3 is used. The generative AI model generates text based on the query and returns the generated results to the server.

[0850] Next, the server compares the search results obtained from the information search service with the results generated by the AI. This comparison is performed by matching the titles and summaries of the search results with the corresponding parts of the generated results. The server then calculates the degree of match. The degree of match is a numerical value that evaluates the degree of match between the search results and the generated results.

[0851] Next, the server performs a scoring process. Based on the results generated by the AI, the server scores the accuracy of each search result. Information accuracy is a numerical value that indicates how accurate and reliable the generated results are.

[0852] Based on these relevance and accuracy scores, the server re-ranks all search results. For example, if a search result has a relevance score of 90 and an accuracy score of 95, the overall score will be calculated as 92.5. Search results with higher scores will ultimately be ranked higher.

[0853] Finally, the server sends the re-ranked search results to the terminal, which then displays the re-ranked search results to the user, allowing the user to quickly check highly reliable and accurate information.

[0854] For example, if a user enters the query "capital of Japan," the server compares the results obtained from the information search service and the generative AI model, and scores search results containing "Tokyo" with high matching and information accuracy. As a result of the re-ranking process, information related to "Tokyo" is displayed at the top.

[0855] An example prompt is:

[0856] Query: capital of japan

[0857] Explanation: What is the capital of Japan?

[0858] By implementing this invention, users can quickly and accurately obtain the information they need. By utilizing the advanced natural language processing capabilities of the generated artificial intelligence model, it is possible to provide more reliable search results than conventional information search services.

[0859] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0860] Step 1:

[0861] User enters a search query

[0862] A user enters a search query into a device. For example, the user enters "capital of Japan."

[0863] Input: Search query (e.g. "Capital of Japan")

[0864] Output: Search queries sent from the device to the internet

[0865] Step 2:

[0866] Device submitting search queries

[0867] The terminal transmits the search query entered by the user to the information search service via the Internet, and also transmits the same query to the server.

[0868] Input: User's search query

[0869] Output: Search query sent to information retrieval service and server

[0870] Step 3:

[0871] Server obtains search results from information search service

[0872] The server retrieves search results from the information search service based on the submitted search query, using an API to retrieve results from a search engine (e.g., a general-purpose search engine).

[0873] Input: Search results from information search services

[0874] Output: Search results saved on the server

[0875] Step 4:

[0876] Server sends queries to the generative AI model

[0877] The server sends the received search query to a generative AI model, such as a natural language processing model (e.g., OpenAI's GPT-3).

[0878] Input: Search query

[0879] Output: Request to the generative AI model

[0880] Step 5:

[0881] Obtaining generated results from the generative AI model by the server

[0882] The generative AI model generates text based on the search query and returns the results to the server, which receives the generated results.

[0883] Input: Response from a generative AI model

[0884] Output: Generated results stored on the server

[0885] Step 6:

[0886] Comparing server search results with generated results

[0887] The server compares the search results from the information retrieval service with the generated results from the generative AI model by matching the titles and summaries of the search results with the corresponding parts of the generated results.

[0888] Input: Search results and generated results

[0889] Data processing: Matching with text comparison algorithms

[0890] Output: Match score for each result

[0891] Step 7:

[0892] Server-based scoring of information accuracy

[0893] The server scores the information accuracy of each search result based on the results generated by the AI. Information accuracy is a numerical value that evaluates the accuracy and reliability of the generated results.

[0894] Input: Generated results

[0895] Data Computation: Information Accuracy Scoring Algorithm

[0896] Output: Information accuracy score for each search result

[0897] Step 8:

[0898] Server Re-ranking

[0899] The server re-ranks all search results based on their relevance and accuracy scores, with more reliable results being ranked higher.

[0900] Input: Match score and information accuracy score

[0901] Data arithmetic: Re-ranking by weighted average of scores

[0902] Output: Re-ranked search results

[0903] Step 9:

[0904] Server sends re-ranked search results

[0905] The server transmits the re-ranked search results to the terminal.

[0906] Input: Reranked search results

[0907] Output: Search results sent to your device

[0908] Step 10:

[0909] Displaying search results by device

[0910] The terminal displays the re-ranked search results to the user, who can then check the displayed search results and quickly and accurately obtain the information they need.

[0911] Input: Reranked search results

[0912] Output: Search results displayed to the user

[0913] (Application example 1)

[0914] 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."

[0915] In conventional food delivery services, even when a user inputs the name of a desired dish or restaurant, the accuracy and reliability of the search results are insufficient, making it difficult to quickly find the best option. Furthermore, the search results often contain advertisements and irrelevant information, hindering the user's ability to select the best restaurant. The objective of the present invention is to solve these problems and quickly and accurately provide the information desired by the user.

[0916] 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.

[0917] In this invention, the server includes means for receiving a search query, means for obtaining search results from a search engine based on the search query, means for transmitting the search query to an AI model that generates the search query and obtaining the generated results, means for comparing the search results with the generated results and calculating a degree of match, means for scoring the information accuracy of the search results based on the generated results, means for re-ranking the search results based on the degree of match and the information accuracy, means for displaying the re-ranked search results on a terminal, and means for presenting food delivery options optimized using the AI ​​model based on the name of a dish or restaurant entered by a user. This enables a user to quickly obtain more reliable and accurate food delivery information in response to a search query.

[0918] A "search query" is a keyword or phrase that a user enters when searching for information.

[0919] A "search engine" is a system that searches for information on the Internet and provides related web pages and data.

[0920] A "generative artificial intelligence model" is an AI algorithm that performs natural language processing based on user input and generates information.

[0921] The "degree of match" is an index for evaluating the degree of match between the search results and the generated results.

[0922] "Information accuracy" is a score that indicates the accuracy and reliability of the information calculated by the generative AI model.

[0923] "Re-ranking" is a process of re-ranking the acquired search results based on the scores of match and information accuracy.

[0924] A "terminal" is an electronic device such as a personal computer, smartphone, or tablet.

[0925] The "dish name" is the name of a specific food item that a user specifies when using a food delivery service.

[0926] "Restaurant name" refers to the name of the restaurant or cafe that provides the food delivery service.

[0927] "Food delivery options" refers to multiple delivery service options offered for the food or restaurant searched for by the user.

[0928] The present invention provides a system for optimizing search results for food delivery services based on a search query entered by a user. The system supports a series of processes including receiving a search query, obtaining search results, obtaining generated results using an artificial intelligence model, comparing and scoring both results, re-ranking, and displaying the final results.

[0929] Hardware and Software Configuration

[0930] Server: A server with powerful processing power is responsible for processing search queries and aggregating results.

[0931] Device: A device such as a personal computer, smartphone, or tablet is used as the device where the user enters the search query and displays the final results.

[0932] software:

[0933] Programming language: Python

[0934] Libraries: Requests library (used to communicate with search engine APIs and food delivery APIs)

[0935] Generative AI models: Leverage artificial intelligence generative models (e.g., GPT-3) through the OpenAI API.

[0936] Data processing and calculation flow

[0937] The server first receives a search query entered by a user on a device, sends the search query to a food delivery search engine, and retrieves relevant search results. In parallel, the same search query is sent to a generative artificial intelligence model, and retrieves generated results.

[0938] The server then compares the search engine results with the results generated by the generative AI model and calculates the degree of match between them. This degree of match is calculated based on the degree of match between the title and snippet of the search result and the title and snippet of the generated result. The information accuracy of the search result is also scored based on the reliability of the generated result.

[0939] The server re-ranks the search results based on the match and information accuracy scores, and the re-ranked search results are sent to the terminal and displayed to the user.

[0940] Specific examples

[0941] For example, if a user types "sushi" into their device, the server retrieves food delivery options related to "sushi" from the search engine and the generative AI model. Then, it compares both results and re-ranks them based on match and information accuracy. Finally, the user is presented with the most reliable and accurate sushi delivery options.

[0942] Prompt Sentence Examples

[0943] "Give me the best food delivery options for Sushi"

[0944] Such a system allows users to obtain reliable food delivery information more efficiently.

[0945] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0946] Step 1:

[0947] A user inputs a search query into a terminal. The input search query is the name of a dish or restaurant that the user is interested in, such as "sushi." The input query becomes the basis for other processing steps in the system.

[0948] Step 2:

[0949] The server receives a search query. The received search query is sent to a search engine API to retrieve relevant search results. For example, if the search query is "sushi," the server retrieves multiple food delivery options that offer sushi. The input is the search query "sushi," and the output is a list of relevant search results.

[0950] Step 3:

[0951] The server sends the same search query to a generative AI model. The generative AI model provides generated results based on the query, for example using the OpenAI API. The input is the search query "sushi," and the output is a list of recommended food delivery options provided by the generative AI. The prompt provided by the generative AI model is "Give me the best food delivery options for Sushi."

[0952] Step 4:

[0953] The server compares the search results with the results of the generative AI model. Specifically, it compares the titles and snippets of the search results with those of the generated results and calculates the degree of match. The degree of match is quantified, for example, to show how many common elements each result contains. The input is the search results and the generated results, and the output is a score indicating the degree of match between each result.

[0954] Step 5:

[0955] The server scores the information accuracy of the search results based on the generated results. In this process, an information accuracy score is assigned to each search result based on the reliability index of the generative AI model and the content of the search result. The input is the match score and the reliability index of the generative AI model, and the output is the information accuracy score for each search result.

[0956] Step 6:

[0957] The server re-ranks the search results based on the relevance and accuracy scores. This re-ranking adjusts the ranking so that results with high relevance and accuracy are displayed at the top. The input is the relevance and accuracy scores of each search result, and the output is a list of re-ranked search results.

[0958] Step 7:

[0959] The server sends the re-ranked search results to the device. The user reviews these results on the device and selects the food delivery option that best suits them. The input is a list of re-ranked search results, and the output is the optimized search results displayed on the device.

[0960] 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.

[0961] The present invention combines a system that utilizes a search engine and a generating artificial intelligence model to optimize search results based on a search query entered by a user, with an emotion engine that recognizes the user's emotions. The system performs the following continuous processes: receiving a search query, obtaining search results, obtaining generated results from the generating AI, comparing and scoring the results, re-ranking, displaying the final results, and filtering and adjusting the search results based on the user's emotions.

[0962] First, a user enters a search query into a device. This query is sent to a search engine on the Internet, and traditional search results are obtained. In parallel, the server sends the same search query to an artificial intelligence model that generates it, and receives the generated results from the AI.

[0963] Next, the server compares the acquired search results with the results generated by the AI. Specifically, it matches the titles and snippets of the search results with the corresponding parts of the generated results to calculate the degree of match. This degree of match serves as an index for evaluating the reliability of each search result. Furthermore, the server scores the information accuracy of each search result based on the generated results. This quantifies how closely the AI's results match the actual search results.

[0964] Based on these relevance and information accuracy scores, the server re-ranks all search results. For example, a search result with a relevance score of 90 and an information accuracy score of 95 will receive a higher overall score of 92.5. In this way, the re-ranked search results are determined.

[0965] Furthermore, this system incorporates an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's emotions through voice input and facial recognition data when the user enters a search query. This emotion information is used to filter search results and adjust the display order. For example, if the user is angry, search results that emphasize information accuracy will be displayed at the top, providing results that will help the user solve their problem. On the other hand, if the user is relaxed, general information and interesting content will be displayed preferentially.

[0966] This allows users to quickly and accurately obtain the most appropriate information according to their emotions. As a specific example, if a user enters the query "capital of Japan," the server will evaluate "Tokyo" as having a high degree of match and information accuracy, and then adjust the display order based on the user's emotions. For example, if the user is in a hurry, it is possible to provide information quickly by displaying search results for "Tokyo" at the top.

[0967] This system combines the advanced natural language processing capabilities of the artificial intelligence model it generates with the user emotion recognition capabilities of the emotion engine to deliver more reliable search results and information that is optimized for users compared to conventional search engines.

[0968] The processing flow will be explained below.

[0969] Step 1:

[0970] The user enters a search query into the device. The user clicks the "Search" button to submit the query.

[0971] Step 2:

[0972] The terminal sends the user's search query to the server.

[0973] Step 3:

[0974] The server receives the user's search query and logs the query.

[0975] Step 4:

[0976] The server sends the received search query to the search engine, and waits for a response from the search engine.

[0977] Step 5:

[0978] The search engine returns the search results to the server, which temporarily stores the received search results in its internal database.

[0979] Step 6:

[0980] The server sends the same search query to the AI ​​model, and the server waits for the AI ​​model to generate the results.

[0981] Step 7:

[0982] The generating AI model sends the generated results back to the server, which stores the received results in its internal database.

[0983] Step 8:

[0984] The server compares the search results it has obtained with the results generated by the AI. Specifically, it matches the title and snippet of the search results with the generated results.

[0985] Step 9:

[0986] The server calculates the degree of match for each search result, which is a numerical representation of the degree of match between the content of the search result and the generated result.

[0987] Step 10:

[0988] The server assigns a score to each search result based on its accuracy, calculated on a scale of 0 to 100.

[0989] Step 11:

[0990] The server re-ranks all search results based on relevance and information accuracy, calculates new rankings, and forms a re-ranked search result list.

[0991] Step 12:

[0992] The device recognizes the emotion expressed when the user enters a query through an emotion engine, and acquires emotion data through voice input and facial recognition using a camera.

[0993] Step 13:

[0994] The emotion engine analyzes the user's emotions and sends the emotion data to the server, where it is stored in a database.

[0995] Step 14:

[0996] The server adjusts the display order and content of search results based on the emotional data. For example, if the user is feeling stressed, it will prioritize displaying information that is more important.

[0997] Step 15:

[0998] The server sends the re-ranked search result list to the terminal.

[0999] Step 16:

[1000] The terminal displays the final search results to the user, who then checks the results and obtains the information they need.

[1001] Step 17:

[1002] Users can click on search results to view details, which allows them to efficiently obtain the most reliable information that best suits their emotions.

[1003] Through these processing steps, the system can provide optimal search results that take into account the user's sentiment and search accuracy. For example, if a user enters the query "capital of Japan," the server will evaluate "Tokyo" as having a high degree of match and information accuracy, and then adjust the display order based on the user's sentiment. For example, if the user is in a hurry, the server can quickly provide information by displaying search results for "Tokyo" at the top.

[1004] Example 2

[1005] 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."

[1006] Conventional search engines have difficulty providing reliable information in response to user-entered search queries, resulting in the inability of users to quickly and accurately obtain the information they need. Furthermore, conventional systems do not display search results that take the user's emotional state into account, making improving the user experience a challenge. Furthermore, there is a demand for more advanced information provision by integrating and utilizing search results with results generated by generative AI models.

[1007] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1008] In this invention, the server includes means for receiving a search query entered by a user, means for obtaining search results from a search technology based on the search query, means for transmitting the search query to an AI model that generates the search query and obtaining the generated results, means for comparing the search results with the generated results and calculating a degree of match, means for scoring the information accuracy of the search results based on the generated results, means for re-ranking the search results based on the degree of match and the information accuracy, means for recognizing and analyzing the user's emotions, means for filtering and adjusting the re-ranked search results based on the emotions, and means for displaying the re-ranked search results on a terminal, thereby enabling a user to quickly and accurately obtain reliable search results that correspond to their emotional state.

[1009] A "search query" refers to a string or phrase that a user enters to obtain information.

[1010] "Search technology" refers to algorithms and systems for retrieving information from the Internet or within databases.

[1011] An "artificial intelligence model" refers to a program or system that performs natural language processing and pattern recognition based on large amounts of data.

[1012] "Generated results" refers to the answers or information generated by an AI model based on an input search query.

[1013] "Matchability" refers to an index that numerically represents how closely the search results match the generated results.

[1014] "Information accuracy" refers to an indicator that evaluates how accurate and reliable search results or generated results are.

[1015] "Reranking" refers to the re-ordering of search results and generated results based on specified criteria.

[1016] An "emotion engine" refers to a system that analyzes emotions based on the user's voice input and facial recognition data.

[1017] "Filtering" refers to the process of sorting and ordering information based on specific criteria.

[1018] The present invention combines a system that utilizes search technology and artificial intelligence models to generate optimized search results based on a user-entered search query with an emotion engine that recognizes the user's emotions. This system performs multiple processing steps in succession to highly optimize the user's search experience.

[1019] First, a user enters a search query into a device. For example, the user enters the query "Today's weather in Tokyo." This input is entered into the search bar and confirmed by pressing the Enter key. Next, the device sends this entered search query to a search technology on the Internet (e.g., a general search engine). At the same time, the server sends the same query to an artificial intelligence model (e.g., a generative AI specialized in natural language processing) that generates it.

[1020] The server receives both the search results returned by the search engine and the generated results provided by the generation AI. For example, the search engine may return 12 search results, and the generation AI may return a sentence-format answer. The server then compares these results by matching the titles and snippets of the search results with the corresponding parts of the generated results and calculating the degree of match. The degree of match is a numerical indicator of how closely the search results and the output of the generation AI match.

[1021] Furthermore, the server scores the information accuracy of each search result based on this degree of match. For example, if a search result has a degree of match of 90 and an information accuracy of 95, its overall score will be (90 + 95) / 2 = 92.5. This allows the server to re-rank all search results. Search results with higher information accuracy will be displayed at the top, allowing users to quickly obtain more reliable information.

[1022] The system also incorporates an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's emotions through voice input and facial recognition data when the user enters a search query. For example, if the user is angry when entering a search query, the emotion engine detects this state and filters the search results to display search results that emphasize information accuracy. In addition, if the user is relaxed, general information and interesting content are prioritized.

[1023] The search results, filtered and adjusted in this way, are finally sent from the server to the device and displayed on the screen, allowing the user to quickly and accurately obtain the most appropriate information according to their emotional state.

[1024] Specific examples

[1025] When a user searches for "events in Tokyo next weekend," the following process occurs:

[1026] 1. The user types "Events in Tokyo next weekend" into the terminal.

[1027] 2. The device sends this query to a search technology (e.g., a general search engine), and the server sends the same query to an artificial intelligence model that generates it (e.g., a generative AI specialized in natural language processing).

[1028] 3. The server receives search results from the search engine and generated results from the generation AI.

[1029] 4. The server compares the received results and scores them based on their matching and accuracy. For example, the matching score is 85 and the accuracy is 90.

[1030] 5. The server re-ranks the search results based on these scores. A search result with a relevance of 85 and information accuracy of 90 will be ranked higher with an overall score of 87.5.

[1031] 6. The device sends the user's voice input and facial recognition data to the emotion engine to determine the user's emotions.

[1032] 7. The server adjusts the search results based on the analysis results of the emotion engine. If the user is relaxed, interesting event information will be displayed at the top.

[1033] 8. Finally, the device displays the filtered search results to the user, allowing them to relax and enjoy information about events in Tokyo next weekend.

[1034] This system allows users to obtain highly reliable and accurate search results in a format that is optimal for their emotional state.

[1035] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1036] Step 1:

[1037] A user enters a search query.

[1038] The user enters the prompt phrase "Today's weather in Tokyo" into the search bar on the device. Once this input is confirmed, the device prepares to send the search query to the Internet.

[1039] Input: The search query entered by the user (e.g., "Today's weather in Tokyo")

[1040] Output: The device is ready to send a search query.

[1041] Step 2:

[1042] Submit queries to search technology and generative AI.

[1043] The device sends the entered search query to the search technology, and at the same time, the server sends the same query to an artificial intelligence model that generates it (for example, an internet search engine is used as the search technology, and a natural language processing model is used as the generation AI).

[1044] Input: The search query entered

[1045] Output: Requests sent to search techniques and generative AI models

[1046] Step 3:

[1047] Receive search results and generated AI results.

[1048] The server receives both the search results returned by the search engine and the generated results provided by the generative AI model. For example, the search engine returns 12 search results and the generative AI returns one generated result.

[1049] Input: Results from search techniques and generative AI models

[1050] Output: Both result sets (e.g. 12 search results from the search engine and 1 generated result from the generative AI)

[1051] Step 4:

[1052] The search results are compared with the results of the generated AI to calculate the degree of match.

[1053] The server compares the title and snippet of the generated result with the received search result and calculates the degree of match. The degree of match is expressed as a numerical value indicating the degree of content match between the search result and the generated result. For example, the degree of match between a specific search result and the generated result is evaluated as 90.

[1054] Input: Search results, generated results

[1055] Output: Matching score (e.g. 90)

[1056] Step 5:

[1057] Scoring the accuracy of information in search results.

[1058] The server assigns a score to each search result based on its relevance. For example, if a search result has a relevance score of 90 and an information accuracy score of 95, its overall score will be (90 + 95) / 2 = 92.5.

[1059] Input: Search results, match rate

[1060] Output: Information accuracy score (e.g. 92.5)

[1061] Step 6:

[1062] Re-rank search results.

[1063] The server re-ranks all search results based on relevance and accuracy. Search results with higher accuracy are ranked higher. For example, a search result with a relevance of 85 and accuracy of 90 would be ranked higher with an overall score of 87.5.

[1064] Input: Information accuracy score

[1065] Output: Re-ranked search results

[1066] Step 7:

[1067] Analyze user sentiment.

[1068] The device sends voice input and facial recognition data when the user enters a search query to the emotion engine, which then analyzes the user's emotions based on this data. For example, it can determine that the user is angry based on their voice tone and facial expression.

[1069] Input: Voice data, facial recognition data

[1070] Output: User's emotional state (e.g. angry)

[1071] Step 8:

[1072] Filter and tailor search results based on sentiment.

[1073] The server reevaluates the search results based on the analysis results of the emotion engine. For example, if the user is angry, the server will filter the search results so that more accurate information is displayed at the top. On the other hand, if the user is relaxed, more interesting information will be displayed at the top.

[1074] Input: Re-ranked search results, user emotional state

[1075] Output: Filtered and refined search results

[1076] Step 9:

[1077] The final result is displayed to the user.

[1078] The server then sends the filtered and adjusted search results to the device, which then displays them on the screen, allowing users to quickly and accurately obtain the most appropriate information based on their emotional state.

[1079] Input: Filtered and refined search results

[1080] Output: Final search results displayed on your device

[1081] As described above, the system performs multiple processing steps in succession to highly optimize the user's search experience.

[1082] (Application example 2)

[1083] 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."

[1084] Modern search engines provide highly accurate search results based on user search queries, but they often fall short of providing optimal information tailored to the user's emotions and circumstances. In particular, online shopping sites require technologies that can efficiently meet user needs, as users' needs vary in terms of the products and information they seek. Conventional search systems are unable to re-rank or filter search results based on the user's emotions and circumstances while ensuring the reliability and accuracy of information, making it difficult to improve the user experience. To address these issues, an improved search system using an emotion engine is needed.

[1085] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1086] In this invention, the server includes means for receiving a search query, means for obtaining search results from a search engine based on the search query, means for transmitting the search query to an artificial intelligence model that generates the search query and obtaining the generated results, means for comparing the search results with the generated results and calculating the degree of match, means for scoring the information accuracy of the search results based on the generated results, means for re-ranking the search results based on the degree of match and the information accuracy, means for displaying the re-ranked search results on a terminal, means for analyzing a user's emotions, and means for filtering the search results or adjusting the display order based on the emotion information. This makes it possible to display optimal search results according to the user's emotions and situation while ensuring the reliability and information accuracy of the search results.

[1087] A "search query" is a set of character strings or phrases that a user enters when searching for information.

[1088] A "search engine" is a system that collects information on the Internet and provides the most appropriate search results in response to a search query.

[1089] A "generative artificial intelligence model" is an algorithm or system that generates text or data using deep learning or natural language processing techniques.

[1090] The "degree of match" is an index that numerically represents how closely the search results match the generated results.

[1091] "Information accuracy" is a measure for evaluating how accurate search results are with respect to the information a user is looking for.

[1092] "Reranking" is the process of reevaluating search results based on specific criteria and updating the display order.

[1093] A "terminal" is a device, such as a smartphone or computer, that a user uses to input information and receive results.

[1094] "Means for analyzing emotions" refers to technologies and systems that identify and analyze a user's emotional state from their input, facial expressions, voice, etc.

[1095] "Filtering" is the process of sorting search results based on specific criteria and removing unnecessary information.

[1096] MODE FOR CARRYING OUT THE INVENTION

[1097] To implement this invention, a system is required in which a server and a user cooperate through a terminal and provide optimal search results based on the user's search query. This system includes the following technologies and processes:

[1098] Server Processing

[1099] The server first receives a search query entered by a user on a device. This query is then sent as a request to a search engine, which retrieves traditional search results. At the same time, the server also sends the same search query to an artificial intelligence model (e.g., GPT-3) that generates it, and receives the generated results.

[1100] Next, the server compares the retrieved search results with the generated results and calculates the degree of match. Specifically, it verifies the title and snippet of the search results with the corresponding parts of the generated results and quantifies the degree of match. This degree of match serves as an index for evaluating the reliability of each search result.

[1101] Furthermore, the server scores each search result based on its generated results, quantifying how closely the generated AI results match the actual search results. Based on these match and accuracy scores, the server re-ranks all search results.

[1102] User sentiment analysis

[1103] When a user enters a search query, the device sends voice input and facial recognition data to the server. The server then uses an emotion engine to analyze the user's emotions. This emotion information is used to filter search results and adjust the display order. For example, if the user is feeling stressed, search results that emphasize information accuracy will be displayed at the top, providing results that will help the user solve their problem. On the other hand, if the user is relaxed, general information and interesting content will be displayed preferentially.

[1104] Specific processing examples

[1105] For example, if a user enters the search query "cheap laptops," the server receives search results from the search engine and results generated by the generative AI model. Then, these results are compared, scored for consistency and information accuracy, and re-ranked. Finally, if the emotion engine determines that the user's emotion is "hurry," the server prioritizes the display of the most highly rated products.

[1106] Examples of prompts that can be used include the following:

[1107] User search query: "cheap laptop"

[1108] Emotion: "I'm in a hurry"

[1109] Search result title: "cheapest laptops 2023"

[1110] Generated AI result snippet: "The cheapest laptops in 2023 are all-in-one..."

[1111] Hardware and software used

[1112] The following hardware and software are used to implement this invention:

[1113] Server: A high-performance computer for data processing and result generation

[1114] Emotion Engine: Software for Analyzing User Emotions

[1115] Generative AI models: Algorithms for natural language processing (e.g., GPT-3)

[1116] Device: A smartphone or computer where users input information and receive results.

[1117] This allows users to quickly and accurately obtain optimal search results based on their own emotions.

[1118] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1119] Step 1:

[1120] A user enters a search query into a device.

[1121] Input: User's search query (e.g. "cheap laptops")

[1122] Output: Search query sent from device to server

[1123] How it works: The user enters the information they want to find into the device's search bar and presses the search button. The device then sends this search query to the server.

[1124] Step 2:

[1125] A server receives a search query and sends it to a search engine to obtain conventional search results.

[1126] Input: Search query

[1127] Output: Search results from a search engine

[1128] Specific operation: The server sends the received search query to the search engine and retrieves search results including relevant indexes and links.

[1129] Step 3:

[1130] The server sends the same search query to the generative AI model and receives the generated results.

[1131] Input: Search query

[1132] Output: The generated results from the generative AI model (e.g., text generated by GPT-3)

[1133] How it works: The server sends search queries to the generative AI model and receives the generated text and data from the model.

[1134] Step 4:

[1135] The server compares the search results it has acquired with the generated results and calculates the degree of match.

[1136] Input: Search results, generated results

[1137] Output: Match score (e.g. title or snippet match)

[1138] What it does: The server matches the title and snippet of the search result with the generated result and calculates a score based on how well it matches.

[1139] Step 5:

[1140] The server scores the accuracy of the information in the search results based on the generated results.

[1141] Input: Search results, generated results

[1142] Output: Information accuracy score

[1143] How it works: The server evaluates how accurate the search results are based on the generated results and assigns an information accuracy score to each result.

[1144] Step 6:

[1145] The server re-ranks the search results based on match and information accuracy.

[1146] Input: Match score, information accuracy score

[1147] Output: Re-ranked search results

[1148] Specific operation: The server re-evaluates the search results based on the degree of match and information accuracy, and ranks them in a new order.

[1149] Step 7:

[1150] The device sends voice input and facial recognition data to a server to analyze the user's emotions.

[1151] Input: User voice data, facial recognition data

[1152] Output: Parsed emotion data

[1153] How it works: The device captures the user's voice and face with a camera and sends the data to a server, where the emotion engine analyzes it to determine the user's emotional state.

[1154] Step 8:

[1155] The server filters or adjusts the display order of search results based on the user's emotional information.

[1156] Input: Reranked search results, sentiment data

[1157] Output: Final search results according to sentiment

[1158] Specific operation: Based on the emotional information analyzed by the emotion engine, the server further filters the search results and displays them in the order that best suits the user's current emotions.

[1159] Step 9:

[1160] The server displays the final search results on the device.

[1161] Input: Final search results according to sentiment

[1162] Output: The final search results displayed on your terminal

[1163] Specific operation: The server sends the filtered and re-ranked search results to the terminal and displays them to the user.

[1164] Examples of prompts:

[1165] User search query: "cheap laptop"

[1166] Emotion: "I'm in a hurry"

[1167] Search result title: "cheapest laptops 2023"

[1168] Generated AI result snippet: "The cheapest laptops in 2023 are all-in-one..."

[1169] 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.

[1170] 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.

[1171] 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.

[1172] [Fourth embodiment]

[1173] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1174] 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.

[1175] 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).

[1176] 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.

[1177] 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.

[1178] 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).

[1179] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1180] 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.

[1181] 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.

[1182] 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.

[1183] 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.

[1184] 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.

[1185] 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."

[1186] The present invention is a system for optimizing search results based on a search query entered by a user using a search engine and a generating artificial intelligence model. The system performs the following consecutive processes: receiving a search query, obtaining search results, obtaining generated results from the generating AI, comparing and scoring the results, re-ranking, and displaying the final results.

[1187] First, a user enters a search query into a device. This query is sent to a search engine on the Internet, and traditional search results are obtained. In parallel, the server sends the same search query to an artificial intelligence model that generates it, and receives the generated results from the AI.

[1188] Next, the server compares the acquired search results with the results generated by the AI. Specifically, it matches the titles and snippets of the search results with the corresponding parts of the generated results to calculate the degree of match. This degree of match serves as an index for evaluating the reliability of each search result. Furthermore, the server scores the information accuracy of each search result based on the generated results. This quantifies how closely the AI's results match the actual search results.

[1189] Based on these relevance and information accuracy scores, the server re-ranks all search results. For example, a search result with a relevance score of 90 and an information accuracy score of 95 will receive a higher overall score of 92.5. In this way, the re-ranked search results are determined.

[1190] Finally, the server sends the re-ranked search results to the device and displays them to the user. This allows users to quickly obtain information with high reliability and guaranteed accuracy. For example, if a user enters the query "capital of Japan," the server will evaluate "Tokyo" as having a high degree of match and information accuracy and display it at the top.

[1191] By utilizing the advanced natural language processing capabilities of the AI ​​model it generates, this system is able to provide more reliable search results than conventional search engines, allowing users to quickly and accurately obtain the information they need.

[1192] The processing flow will be explained below.

[1193] Step 1:

[1194] The user enters a search query into the device. The user clicks the "Search" button to submit the query.

[1195] Step 2:

[1196] The server receives the user's search query and logs the query.

[1197] Step 3:

[1198] The server sends the received search query to the search engine, and waits for a response from the search engine.

[1199] Step 4:

[1200] The search engine returns the search results to the server, which temporarily stores the received search results in its internal database.

[1201] Step 5:

[1202] The server sends the same search query to the AI ​​model, and the server waits for the AI ​​model to generate the results.

[1203] Step 6:

[1204] The generating AI model sends the generated results back to the server, which stores the received results in its internal database.

[1205] Step 7:

[1206] The search results obtained by the server are compared with the results generated by the AI. The comparison mainly focuses on the titles and snippets of the search results.

[1207] Step 8:

[1208] The server calculates a match score for each search result, which involves quantifying the match between the content of the search result and the generated result.

[1209] Step 9:

[1210] The server scores each search result based on its accuracy, which is based on how reliable the generated result is.

[1211] Step 10:

[1212] The server re-ranks all search results based on relevance and information accuracy, calculates new rankings, and forms a re-ranked search result list.

[1213] Step 11:

[1214] The server sends the re-ranked search result list to the terminal, and the terminal displays the received search results for the user to review.

[1215] Step 12:

[1216] The user checks the displayed search results and obtains the information they need. The user clicks on the search results to view the details.

[1217] Example 1

[1218] 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."

[1219] Modern information retrieval systems often provide irrelevant or unreliable information in response to user-entered search queries. This prevents users from quickly obtaining the information they need, reducing search efficiency. Furthermore, traditional information retrieval services lack the means to adequately evaluate the reliability of search results, exposing users to the risk of making decisions based on incorrect information.

[1220] 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.

[1221] In this invention, the server includes means for receiving a search query, means for obtaining search results from an information search service, and means for sending a query to a generating artificial intelligence model and obtaining the generated results, thereby making it possible to calculate the degree of match between the obtained search results and the generated results, score the information accuracy based on that, and provide the user with re-ranked search results.

[1222] A "search query" is a string of characters that a user enters to search for information.

[1223] An "information search service" is a system or engine for obtaining information on the Internet, generally referring to a search engine.

[1224] A "generative artificial intelligence model" is a machine learning model for generating text based on a given query.

[1225] The "generation result" is the text data that the generating artificial intelligence model outputs in response to the input query.

[1226] "Search results" refers to the relevant information returned by an information retrieval service in response to a search query.

[1227] The "degree of match" is a numerical value or index that evaluates the degree of match between the search result and the generated result.

[1228] "Information accuracy" is a numerical value or index that indicates how accurate and reliable the generated results are.

[1229] "Reranking" is the process of reevaluating and rearranging the ranking of search results based on relevance and information accuracy.

[1230] "Terminal" refers to the device on which a user enters a search query and displays results, including a PC or smartphone.

[1231] This invention is a system for optimizing search results based on a user-provided search query by utilizing an information search service and a generative AI model. The system performs the following consecutive processes: receiving a search query, obtaining search results, obtaining generated results from a generative AI model, comparing the results and calculating the degree of match, scoring the information accuracy, re-ranking, and displaying the final results.

[1232] First, a user inputs a search query into a terminal. For example, the user inputs the query "capital of Japan." This query is sent to an information search service (e.g., a general-purpose search engine) via the Internet. At the same time, the same query is sent to a server.

[1233] The server sends the received search query to a generative AI model (e.g., a natural language processing model). Specifically, OpenAI's GPT-3 is used. The generative AI model generates text based on the query and returns the generated results to the server.

[1234] Next, the server compares the search results obtained from the information search service with the results generated by the AI. This comparison is performed by matching the titles and summaries of the search results with the corresponding parts of the generated results. The server then calculates the degree of match. The degree of match is a numerical value that evaluates the degree of match between the search results and the generated results.

[1235] Next, the server performs a scoring process. Based on the results generated by the AI, the server scores the accuracy of each search result. Information accuracy is a numerical value that indicates how accurate and reliable the generated results are.

[1236] Based on these relevance and accuracy scores, the server re-ranks all search results. For example, if a search result has a relevance score of 90 and an accuracy score of 95, the overall score will be calculated as 92.5. Search results with higher scores will ultimately be ranked higher.

[1237] Finally, the server sends the re-ranked search results to the terminal, which then displays the re-ranked search results to the user, allowing the user to quickly check highly reliable and accurate information.

[1238] For example, if a user enters the query "capital of Japan," the server compares the results obtained from the information search service and the generative AI model, and scores search results containing "Tokyo" with high matching and information accuracy. As a result of the re-ranking process, information related to "Tokyo" is displayed at the top.

[1239] An example prompt is:

[1240] Query: capital of japan

[1241] Explanation: What is the capital of Japan?

[1242] By implementing this invention, users can quickly and accurately obtain the information they need. By utilizing the advanced natural language processing capabilities of the generated artificial intelligence model, it is possible to provide more reliable search results than conventional information search services.

[1243] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1244] Step 1:

[1245] User enters a search query

[1246] A user enters a search query into a device. For example, the user enters "capital of Japan."

[1247] Input: Search query (e.g. "Capital of Japan")

[1248] Output: Search queries sent from the device to the internet

[1249] Step 2:

[1250] Device submitting search queries

[1251] The terminal transmits the search query entered by the user to the information search service via the Internet, and also transmits the same query to the server.

[1252] Input: User's search query

[1253] Output: Search query sent to information retrieval service and server

[1254] Step 3:

[1255] Server obtains search results from information search service

[1256] The server retrieves search results from the information search service based on the submitted search query, using an API to retrieve results from a search engine (e.g., a general-purpose search engine).

[1257] Input: Search results from information search services

[1258] Output: Search results saved on the server

[1259] Step 4:

[1260] Server sends queries to the generative AI model

[1261] The server sends the received search query to a generative AI model, such as a natural language processing model (e.g., OpenAI's GPT-3).

[1262] Input: Search query

[1263] Output: Request to the generative AI model

[1264] Step 5:

[1265] Obtaining generated results from the generative AI model by the server

[1266] The generative AI model generates text based on the search query and returns the results to the server, which receives the generated results.

[1267] Input: Response from a generative AI model

[1268] Output: Generated results stored on the server

[1269] Step 6:

[1270] Comparing server search results with generated results

[1271] The server compares the search results from the information retrieval service with the generated results from the generative AI model by matching the titles and summaries of the search results with the corresponding parts of the generated results.

[1272] Input: Search results and generated results

[1273] Data processing: Matching with text comparison algorithms

[1274] Output: Match score for each result

[1275] Step 7:

[1276] Server-based scoring of information accuracy

[1277] The server scores the information accuracy of each search result based on the results generated by the AI. Information accuracy is a numerical value that evaluates the accuracy and reliability of the generated results.

[1278] Input: Generated results

[1279] Data Computation: Information Accuracy Scoring Algorithm

[1280] Output: Information accuracy score for each search result

[1281] Step 8:

[1282] Server Re-ranking

[1283] The server re-ranks all search results based on their relevance and accuracy scores, with more reliable results being ranked higher.

[1284] Input: Match score and information accuracy score

[1285] Data arithmetic: Re-ranking by weighted average of scores

[1286] Output: Re-ranked search results

[1287] Step 9:

[1288] Server sends re-ranked search results

[1289] The server transmits the re-ranked search results to the terminal.

[1290] Input: Reranked search results

[1291] Output: Search results sent to your device

[1292] Step 10:

[1293] Displaying search results by device

[1294] The terminal displays the re-ranked search results to the user, who can then check the displayed search results and quickly and accurately obtain the information they need.

[1295] Input: Reranked search results

[1296] Output: Search results displayed to the user

[1297] (Application example 1)

[1298] 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."

[1299] In conventional food delivery services, even when a user inputs the name of a desired dish or restaurant, the accuracy and reliability of the search results are insufficient, making it difficult to quickly find the best option. Furthermore, the search results often contain advertisements and irrelevant information, hindering the user's ability to select the best restaurant. The objective of the present invention is to solve these problems and quickly and accurately provide the information desired by the user.

[1300] 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.

[1301] In this invention, the server includes means for receiving a search query, means for obtaining search results from a search engine based on the search query, means for transmitting the search query to an AI model that generates the search query and obtaining the generated results, means for comparing the search results with the generated results and calculating a degree of match, means for scoring the information accuracy of the search results based on the generated results, means for re-ranking the search results based on the degree of match and the information accuracy, means for displaying the re-ranked search results on a terminal, and means for presenting food delivery options optimized using the AI ​​model based on the name of a dish or restaurant entered by a user. This enables a user to quickly obtain more reliable and accurate food delivery information in response to a search query.

[1302] A "search query" is a keyword or phrase that a user enters when searching for information.

[1303] A "search engine" is a system that searches for information on the Internet and provides related web pages and data.

[1304] A "generative artificial intelligence model" is an AI algorithm that performs natural language processing based on user input and generates information.

[1305] The "degree of match" is an index for evaluating the degree of match between the search results and the generated results.

[1306] "Information accuracy" is a score that indicates the accuracy and reliability of the information calculated by the generative AI model.

[1307] "Re-ranking" is a process of re-ranking the acquired search results based on the scores of match and information accuracy.

[1308] A "terminal" is an electronic device such as a personal computer, smartphone, or tablet.

[1309] The "dish name" is the name of a specific food item that a user specifies when using a food delivery service.

[1310] "Restaurant name" refers to the name of the restaurant or cafe that provides the food delivery service.

[1311] "Food delivery options" refers to multiple delivery service options offered for the food or restaurant searched for by the user.

[1312] The present invention provides a system for optimizing search results for food delivery services based on a search query entered by a user. The system supports a series of processes including receiving a search query, obtaining search results, obtaining generated results using an artificial intelligence model, comparing and scoring both results, re-ranking, and displaying the final results.

[1313] Hardware and Software Configuration

[1314] Server: A server with powerful processing power is responsible for processing search queries and aggregating results.

[1315] Device: A device such as a personal computer, smartphone, or tablet is used as the device where the user enters the search query and displays the final results.

[1316] software:

[1317] Programming language: Python

[1318] Libraries: Requests library (used to communicate with search engine APIs and food delivery APIs)

[1319] Generative AI models: Leverage artificial intelligence generative models (e.g., GPT-3) through the OpenAI API.

[1320] Data processing and calculation flow

[1321] The server first receives a search query entered by a user on a device, sends the search query to a food delivery search engine, and retrieves relevant search results. In parallel, the same search query is sent to a generative artificial intelligence model, and retrieves generated results.

[1322] The server then compares the search engine results with the results generated by the generative AI model and calculates the degree of match between them. This degree of match is calculated based on the degree of match between the title and snippet of the search result and the title and snippet of the generated result. The information accuracy of the search result is also scored based on the reliability of the generated result.

[1323] The server re-ranks the search results based on the match and information accuracy scores, and the re-ranked search results are sent to the terminal and displayed to the user.

[1324] Specific examples

[1325] For example, if a user types "sushi" into their device, the server retrieves food delivery options related to "sushi" from the search engine and the generative AI model. Then, it compares both results and re-ranks them based on match and information accuracy. Finally, the user is presented with the most reliable and accurate sushi delivery options.

[1326] Prompt Sentence Examples

[1327] "Give me the best food delivery options for Sushi"

[1328] Such a system allows users to obtain reliable food delivery information more efficiently.

[1329] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1330] Step 1:

[1331] A user inputs a search query into a terminal. The input search query is the name of a dish or restaurant that the user is interested in, such as "sushi." The input query becomes the basis for other processing steps in the system.

[1332] Step 2:

[1333] The server receives a search query. The received search query is sent to a search engine API to retrieve relevant search results. For example, if the search query is "sushi," the server retrieves multiple food delivery options that offer sushi. The input is the search query "sushi," and the output is a list of relevant search results.

[1334] Step 3:

[1335] The server sends the same search query to a generative AI model. The generative AI model provides generated results based on the query, for example using the OpenAI API. The input is the search query "sushi," and the output is a list of recommended food delivery options provided by the generative AI. The prompt provided by the generative AI model is "Give me the best food delivery options for Sushi."

[1336] Step 4:

[1337] The server compares the search results with the results of the generative AI model. Specifically, it compares the titles and snippets of the search results with those of the generated results and calculates the degree of match. The degree of match is quantified, for example, to show how many common elements each result contains. The input is the search results and the generated results, and the output is a score indicating the degree of match between each result.

[1338] Step 5:

[1339] The server scores the information accuracy of the search results based on the generated results. In this process, an information accuracy score is assigned to each search result based on the reliability index of the generative AI model and the content of the search result. The input is the match score and the reliability index of the generative AI model, and the output is the information accuracy score for each search result.

[1340] Step 6:

[1341] The server re-ranks the search results based on the relevance and accuracy scores. This re-ranking adjusts the ranking so that results with high relevance and accuracy are displayed at the top. The input is the relevance and accuracy scores of each search result, and the output is a list of re-ranked search results.

[1342] Step 7:

[1343] The server sends the re-ranked search results to the device. The user reviews these results on the device and selects the food delivery option that best suits them. The input is a list of re-ranked search results, and the output is the optimized search results displayed on the device.

[1344] 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.

[1345] The present invention combines a system that utilizes a search engine and a generating artificial intelligence model to optimize search results based on a search query entered by a user, with an emotion engine that recognizes the user's emotions. The system performs the following continuous processes: receiving a search query, obtaining search results, obtaining generated results from the generating AI, comparing and scoring the results, re-ranking, displaying the final results, and filtering and adjusting the search results based on the user's emotions.

[1346] First, a user enters a search query into a device. This query is sent to a search engine on the Internet, and traditional search results are obtained. In parallel, the server sends the same search query to an artificial intelligence model that generates it, and receives the generated results from the AI.

[1347] Next, the server compares the acquired search results with the results generated by the AI. Specifically, it matches the titles and snippets of the search results with the corresponding parts of the generated results to calculate the degree of match. This degree of match serves as an index for evaluating the reliability of each search result. Furthermore, the server scores the information accuracy of each search result based on the generated results. This quantifies how closely the AI's results match the actual search results.

[1348] Based on these relevance and information accuracy scores, the server re-ranks all search results. For example, a search result with a relevance score of 90 and an information accuracy score of 95 will receive a higher overall score of 92.5. In this way, the re-ranked search results are determined.

[1349] Furthermore, this system incorporates an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's emotions through voice input and facial recognition data when the user enters a search query. This emotion information is used to filter search results and adjust the display order. For example, if the user is angry, search results that emphasize information accuracy will be displayed at the top, providing results that will help the user solve their problem. On the other hand, if the user is relaxed, general information and interesting content will be displayed preferentially.

[1350] This allows users to quickly and accurately obtain the most appropriate information according to their emotions. As a specific example, if a user enters the query "capital of Japan," the server will evaluate "Tokyo" as having a high degree of match and information accuracy, and then adjust the display order based on the user's emotions. For example, if the user is in a hurry, it is possible to provide information quickly by displaying search results for "Tokyo" at the top.

[1351] This system combines the advanced natural language processing capabilities of the artificial intelligence model it generates with the user emotion recognition capabilities of the emotion engine to deliver more reliable search results and information that is optimized for users compared to conventional search engines.

[1352] The processing flow will be explained below.

[1353] Step 1:

[1354] The user enters a search query into the device. The user clicks the "Search" button to submit the query.

[1355] Step 2:

[1356] The terminal sends the user's search query to the server.

[1357] Step 3:

[1358] The server receives the user's search query and logs the query.

[1359] Step 4:

[1360] The server sends the received search query to the search engine, and waits for a response from the search engine.

[1361] Step 5:

[1362] The search engine returns the search results to the server, which temporarily stores the received search results in its internal database.

[1363] Step 6:

[1364] The server sends the same search query to the AI ​​model, and the server waits for the AI ​​model to generate the results.

[1365] Step 7:

[1366] The generating AI model sends the generated results back to the server, which stores the received results in its internal database.

[1367] Step 8:

[1368] The server compares the search results it has obtained with the results generated by the AI. Specifically, it matches the title and snippet of the search results with the generated results.

[1369] Step 9:

[1370] The server calculates the degree of match for each search result, which is a numerical representation of the degree of match between the content of the search result and the generated result.

[1371] Step 10:

[1372] The server assigns a score to each search result based on its accuracy, calculated on a scale of 0 to 100.

[1373] Step 11:

[1374] The server re-ranks all search results based on relevance and information accuracy, calculates new rankings, and forms a re-ranked search result list.

[1375] Step 12:

[1376] The device recognizes the emotion expressed when the user enters a query through an emotion engine, and acquires emotion data through voice input and facial recognition using a camera.

[1377] Step 13:

[1378] The emotion engine analyzes the user's emotions and sends the emotion data to the server, where it is stored in a database.

[1379] Step 14:

[1380] The server adjusts the display order and content of search results based on the emotional data. For example, if the user is feeling stressed, it will prioritize displaying information that is more important.

[1381] Step 15:

[1382] The server sends the re-ranked search result list to the terminal.

[1383] Step 16:

[1384] The terminal displays the final search results to the user, who then checks the results and obtains the information they need.

[1385] Step 17:

[1386] Users can click on search results to view details, which allows them to efficiently obtain the most reliable information that best suits their emotions.

[1387] Through these processing steps, the system can provide optimal search results that take into account the user's sentiment and search accuracy. For example, if a user enters the query "capital of Japan," the server will evaluate "Tokyo" as having a high degree of match and information accuracy, and then adjust the display order based on the user's sentiment. For example, if the user is in a hurry, the server can quickly provide information by displaying search results for "Tokyo" at the top.

[1388] Example 2

[1389] 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."

[1390] Conventional search engines have difficulty providing reliable information in response to user-entered search queries, resulting in the inability of users to quickly and accurately obtain the information they need. Furthermore, conventional systems do not display search results that take the user's emotional state into account, making improving the user experience a challenge. Furthermore, there is a demand for more advanced information provision by integrating and utilizing search results with results generated by generative AI models.

[1391] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1392] In this invention, the server includes means for receiving a search query entered by a user, means for obtaining search results from a search technology based on the search query, means for transmitting the search query to an AI model that generates the search query and obtaining the generated results, means for comparing the search results with the generated results and calculating a degree of match, means for scoring the information accuracy of the search results based on the generated results, means for re-ranking the search results based on the degree of match and the information accuracy, means for recognizing and analyzing the user's emotions, means for filtering and adjusting the re-ranked search results based on the emotions, and means for displaying the re-ranked search results on a terminal, thereby enabling a user to quickly and accurately obtain reliable search results that correspond to their emotional state.

[1393] A "search query" refers to a string or phrase that a user enters to obtain information.

[1394] "Search technology" refers to algorithms and systems for retrieving information from the Internet or within databases.

[1395] An "artificial intelligence model" refers to a program or system that performs natural language processing and pattern recognition based on large amounts of data.

[1396] "Generated results" refers to the answers or information generated by an AI model based on an input search query.

[1397] "Matchability" refers to an index that numerically represents how closely the search results match the generated results.

[1398] "Information accuracy" refers to an indicator that evaluates how accurate and reliable search results or generated results are.

[1399] "Reranking" refers to the re-ordering of search results and generated results based on specified criteria.

[1400] An "emotion engine" refers to a system that analyzes emotions based on the user's voice input and facial recognition data.

[1401] "Filtering" refers to the process of sorting and ordering information based on specific criteria.

[1402] The present invention combines a system that utilizes search technology and artificial intelligence models to generate optimized search results based on a user-entered search query with an emotion engine that recognizes the user's emotions. This system performs multiple processing steps in succession to highly optimize the user's search experience.

[1403] First, a user enters a search query into a device. For example, the user enters the query "Today's weather in Tokyo." This input is entered into the search bar and confirmed by pressing the Enter key. Next, the device sends this entered search query to a search technology on the Internet (e.g., a general search engine). At the same time, the server sends the same query to an artificial intelligence model (e.g., a generative AI specialized in natural language processing) that generates it.

[1404] The server receives both the search results returned by the search engine and the generated results provided by the generation AI. For example, the search engine may return 12 search results, and the generation AI may return a sentence-format answer. The server then compares these results by matching the titles and snippets of the search results with the corresponding parts of the generated results and calculating the degree of match. The degree of match is a numerical indicator of how closely the search results and the output of the generation AI match.

[1405] Furthermore, the server scores the information accuracy of each search result based on this degree of match. For example, if a search result has a degree of match of 90 and an information accuracy of 95, its overall score will be (90 + 95) / 2 = 92.5. This allows the server to re-rank all search results. Search results with higher information accuracy will be displayed at the top, allowing users to quickly obtain more reliable information.

[1406] The system also incorporates an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's emotions through voice input and facial recognition data when the user enters a search query. For example, if the user is angry when entering a search query, the emotion engine detects this state and filters the search results to display search results that emphasize information accuracy. In addition, if the user is relaxed, general information and interesting content are prioritized.

[1407] The search results, filtered and adjusted in this way, are finally sent from the server to the device and displayed on the screen, allowing the user to quickly and accurately obtain the most appropriate information according to their emotional state.

[1408] Specific examples

[1409] When a user searches for "events in Tokyo next weekend," the following process occurs:

[1410] 1. The user types "Events in Tokyo next weekend" into the terminal.

[1411] 2. The device sends this query to a search technology (e.g., a general search engine), and the server sends the same query to an artificial intelligence model that generates it (e.g., a generative AI specialized in natural language processing).

[1412] 3. The server receives search results from the search engine and generated results from the generation AI.

[1413] 4. The server compares the received results and scores them based on their matching and accuracy. For example, the matching score is 85 and the accuracy is 90.

[1414] 5. The server re-ranks the search results based on these scores. A search result with a relevance of 85 and information accuracy of 90 will be ranked higher with an overall score of 87.5.

[1415] 6. The device sends the user's voice input and facial recognition data to the emotion engine to determine the user's emotions.

[1416] 7. The server adjusts the search results based on the analysis results of the emotion engine. If the user is relaxed, interesting event information will be displayed at the top.

[1417] 8. Finally, the device displays the filtered search results to the user, allowing them to relax and enjoy information about events in Tokyo next weekend.

[1418] This system allows users to obtain highly reliable and accurate search results in a format that is optimal for their emotional state.

[1419] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1420] Step 1:

[1421] A user enters a search query.

[1422] The user enters the prompt phrase "Today's weather in Tokyo" into the search bar on the device. Once this input is confirmed, the device prepares to send the search query to the Internet.

[1423] Input: The search query entered by the user (e.g., "Today's weather in Tokyo")

[1424] Output: The device is ready to send a search query.

[1425] Step 2:

[1426] Submit queries to search technology and generative AI.

[1427] The device sends the entered search query to the search technology, and at the same time, the server sends the same query to an artificial intelligence model that generates it (for example, an internet search engine is used as the search technology, and a natural language processing model is used as the generation AI).

[1428] Input: The search query entered

[1429] Output: Requests sent to search techniques and generative AI models

[1430] Step 3:

[1431] Receive search results and generated AI results.

[1432] The server receives both the search results returned by the search engine and the generated results provided by the generative AI model. For example, the search engine returns 12 search results and the generative AI returns one generated result.

[1433] Input: Results from search techniques and generative AI models

[1434] Output: Both result sets (e.g. 12 search results from the search engine and 1 generated result from the generative AI)

[1435] Step 4:

[1436] The search results are compared with the results of the generated AI to calculate the degree of match.

[1437] The server compares the title and snippet of the generated result with the received search result and calculates the degree of match. The degree of match is expressed as a numerical value indicating the degree of content match between the search result and the generated result. For example, the degree of match between a specific search result and the generated result is evaluated as 90.

[1438] Input: Search results, generated results

[1439] Output: Matching score (e.g. 90)

[1440] Step 5:

[1441] Scoring the accuracy of information in search results.

[1442] The server assigns a score to each search result based on its relevance. For example, if a search result has a relevance score of 90 and an information accuracy score of 95, its overall score will be (90 + 95) / 2 = 92.5.

[1443] Input: Search results, match rate

[1444] Output: Information accuracy score (e.g. 92.5)

[1445] Step 6:

[1446] Re-rank search results.

[1447] The server re-ranks all search results based on relevance and accuracy. Search results with higher accuracy are ranked higher. For example, a search result with a relevance of 85 and accuracy of 90 would be ranked higher with an overall score of 87.5.

[1448] Input: Information accuracy score

[1449] Output: Re-ranked search results

[1450] Step 7:

[1451] Analyze user sentiment.

[1452] The device sends voice input and facial recognition data when the user enters a search query to the emotion engine, which then analyzes the user's emotions based on this data. For example, it can determine that the user is angry based on their voice tone and facial expression.

[1453] Input: Voice data, facial recognition data

[1454] Output: User's emotional state (e.g. angry)

[1455] Step 8:

[1456] Filter and tailor search results based on sentiment.

[1457] The server reevaluates the search results based on the analysis results of the emotion engine. For example, if the user is angry, the server will filter the search results so that more accurate information is displayed at the top. On the other hand, if the user is relaxed, more interesting information will be displayed at the top.

[1458] Input: Re-ranked search results, user emotional state

[1459] Output: Filtered and refined search results

[1460] Step 9:

[1461] The final result is displayed to the user.

[1462] The server then sends the filtered and adjusted search results to the device, which then displays them on the screen, allowing users to quickly and accurately obtain the most appropriate information based on their emotional state.

[1463] Input: Filtered and refined search results

[1464] Output: Final search results displayed on your device

[1465] As described above, the system performs multiple processing steps in succession to highly optimize the user's search experience.

[1466] (Application example 2)

[1467] 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."

[1468] Modern search engines provide highly accurate search results based on user search queries, but they often fall short of providing optimal information tailored to the user's emotions and circumstances. In particular, online shopping sites require technologies that can efficiently meet user needs, as users' needs vary in terms of the products and information they seek. Conventional search systems are unable to re-rank or filter search results based on the user's emotions and circumstances while ensuring the reliability and accuracy of information, making it difficult to improve the user experience. To address these issues, an improved search system using an emotion engine is needed.

[1469] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1470] In this invention, the server includes means for receiving a search query, means for obtaining search results from a search engine based on the search query, means for transmitting the search query to an artificial intelligence model that generates the search query and obtaining the generated results, means for comparing the search results with the generated results and calculating the degree of match, means for scoring the information accuracy of the search results based on the generated results, means for re-ranking the search results based on the degree of match and the information accuracy, means for displaying the re-ranked search results on a terminal, means for analyzing a user's emotions, and means for filtering the search results or adjusting the display order based on the emotion information. This makes it possible to display optimal search results according to the user's emotions and situation while ensuring the reliability and information accuracy of the search results.

[1471] A "search query" is a set of character strings or phrases that a user enters when searching for information.

[1472] A "search engine" is a system that collects information on the Internet and provides the most appropriate search results in response to a search query.

[1473] A "generative artificial intelligence model" is an algorithm or system that generates text or data using deep learning or natural language processing techniques.

[1474] The "degree of match" is an index that numerically represents how closely the search results match the generated results.

[1475] "Information accuracy" is a measure for evaluating how accurate search results are with respect to the information a user is looking for.

[1476] "Reranking" is the process of reevaluating search results based on specific criteria and updating the display order.

[1477] A "terminal" is a device, such as a smartphone or computer, that a user uses to input information and receive results.

[1478] "Means for analyzing emotions" refers to technologies and systems that identify and analyze a user's emotional state from their input, facial expressions, voice, etc.

[1479] "Filtering" is the process of sorting search results based on specific criteria and removing unnecessary information.

[1480] MODE FOR CARRYING OUT THE INVENTION

[1481] To implement this invention, a system is required in which a server and a user cooperate through a terminal and provide optimal search results based on the user's search query. This system includes the following technologies and processes:

[1482] Server Processing

[1483] The server first receives a search query entered by a user on a device. This query is then sent as a request to a search engine, which retrieves traditional search results. At the same time, the server also sends the same search query to an artificial intelligence model (e.g., GPT-3) that generates it, and receives the generated results.

[1484] Next, the server compares the retrieved search results with the generated results and calculates the degree of match. Specifically, it verifies the title and snippet of the search results with the corresponding parts of the generated results and quantifies the degree of match. This degree of match serves as an index for evaluating the reliability of each search result.

[1485] Furthermore, the server scores each search result based on its generated results, quantifying how closely the generated AI results match the actual search results. Based on these match and accuracy scores, the server re-ranks all search results.

[1486] User sentiment analysis

[1487] When a user enters a search query, the device sends voice input and facial recognition data to the server. The server then uses an emotion engine to analyze the user's emotions. This emotion information is used to filter search results and adjust the display order. For example, if the user is feeling stressed, search results that emphasize information accuracy will be displayed at the top, providing results that will help the user solve their problem. On the other hand, if the user is relaxed, general information and interesting content will be displayed preferentially.

[1488] Specific processing examples

[1489] For example, if a user enters the search query "cheap laptops," the server receives search results from the search engine and results generated by the generative AI model. Then, these results are compared, scored for consistency and information accuracy, and re-ranked. Finally, if the emotion engine determines that the user's emotion is "hurry," the server prioritizes the display of the most highly rated products.

[1490] Examples of prompts that can be used include the following:

[1491] User search query: "cheap laptop"

[1492] Emotion: "I'm in a hurry"

[1493] Search result title: "cheapest laptops 2023"

[1494] Generated AI result snippet: "The cheapest laptops in 2023 are all-in-one..."

[1495] Hardware and software used

[1496] The following hardware and software are used to implement this invention:

[1497] Server: A high-performance computer for data processing and result generation

[1498] Emotion Engine: Software for Analyzing User Emotions

[1499] Generative AI models: Algorithms for natural language processing (e.g., GPT-3)

[1500] Device: A smartphone or computer where users input information and receive results.

[1501] This allows users to quickly and accurately obtain optimal search results based on their own emotions.

[1502] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1503] Step 1:

[1504] A user enters a search query into a device.

[1505] Input: User's search query (e.g. "cheap laptops")

[1506] Output: Search query sent from device to server

[1507] How it works: The user enters the information they want to find into the device's search bar and presses the search button. The device then sends this search query to the server.

[1508] Step 2:

[1509] A server receives a search query and sends it to a search engine to obtain conventional search results.

[1510] Input: Search query

[1511] Output: Search results from a search engine

[1512] Specific operation: The server sends the received search query to the search engine and retrieves search results including relevant indexes and links.

[1513] Step 3:

[1514] The server sends the same search query to the generative AI model and receives the generated results.

[1515] Input: Search query

[1516] Output: The generated results from the generative AI model (e.g., text generated by GPT-3)

[1517] How it works: The server sends search queries to the generative AI model and receives the generated text and data from the model.

[1518] Step 4:

[1519] The server compares the search results it has acquired with the generated results and calculates the degree of match.

[1520] Input: Search results, generated results

[1521] Output: Match score (e.g. title or snippet match)

[1522] What it does: The server matches the title and snippet of the search result with the generated result and calculates a score based on how well it matches.

[1523] Step 5:

[1524] The server scores the accuracy of the information in the search results based on the generated results.

[1525] Input: Search results, generated results

[1526] Output: Information accuracy score

[1527] How it works: The server evaluates how accurate the search results are based on the generated results and assigns an information accuracy score to each result.

[1528] Step 6:

[1529] The server re-ranks the search results based on match and information accuracy.

[1530] Input: Match score, information accuracy score

[1531] Output: Re-ranked search results

[1532] Specific operation: The server re-evaluates the search results based on the degree of match and information accuracy, and ranks them in a new order.

[1533] Step 7:

[1534] The device sends voice input and facial recognition data to a server to analyze the user's emotions.

[1535] Input: User voice data, facial recognition data

[1536] Output: Parsed emotion data

[1537] How it works: The device captures the user's voice and face with a camera and sends the data to a server, where the emotion engine analyzes it to determine the user's emotional state.

[1538] Step 8:

[1539] The server filters or adjusts the display order of search results based on the user's emotional information.

[1540] Input: Reranked search results, sentiment data

[1541] Output: Final search results according to sentiment

[1542] Specific operation: Based on the emotional information analyzed by the emotion engine, the server further filters the search results and displays them in the order that best suits the user's current emotions.

[1543] Step 9:

[1544] The server displays the final search results on the device.

[1545] Input: Final search results according to sentiment

[1546] Output: The final search results displayed on your terminal

[1547] Specific operation: The server sends the filtered and re-ranked search results to the terminal and displays them to the user.

[1548] Examples of prompts:

[1549] User search query: "cheap laptop"

[1550] Emotion: "I'm in a hurry"

[1551] Search result title: "cheapest laptops 2023"

[1552] Generated AI result snippet: "The cheapest laptops in 2023 are all-in-one..."

[1553] 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.

[1554] 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.

[1555] 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.

[1556] 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.

[1557] 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.

[1558] 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.

[1559] 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).

[1560] 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.

[1561] 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."

[1562] 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.

[1563] 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).

[1564] 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.

[1565] 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.

[1566] 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.

[1567] 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.

[1568] 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.

[1569] 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.

[1570] 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.

[1571] 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.

[1572] 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.

[1573] 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.

[1574] The following is further disclosed regarding the above embodiment.

[1575] (Claim 1)

[1576] means for receiving a search query;

[1577] means for obtaining search results from a search engine based on the search query;

[1578] means for transmitting the search query to an artificial intelligence model for generating the search query and obtaining the generated result;

[1579] means for comparing the search results with the generated results and calculating the degree of match;

[1580] A means for scoring the information accuracy of the search results based on the generated results;

[1581] means for re-ranking the search results based on the degree of match and the information accuracy;

[1582] means for displaying the re-ranked search results on a terminal;

[1583] A system including:

[1584] (Claim 2)

[1585] 2. The system of claim 1, wherein the degree of match is calculated based on the degree of match between the title and snippet of the search result and the title and snippet of the generated result.

[1586] (Claim 3)

[1587] 2. The system of claim 1, wherein the comparison and scoring of the search results and the generated results is performed using a scoring engine based on the reliability of the generated results.

[1588] "Example 1"

[1589] (Claim 1)

[1590] means for receiving a search query;

[1591] means for obtaining search results from an information retrieval service based on the search query;

[1592] means for transmitting the search query to an artificial intelligence model for generating the search query and obtaining the generated result;

[1593] means for comparing the search results with the generated results and calculating the degree of match;

[1594] A means for scoring the information accuracy of the search results based on the generated results;

[1595] means for re-ranking the search results based on the degree of match and the information accuracy;

[1596] means for displaying the re-ranked search results on a terminal;

[1597] A system including:

[1598] (Claim 2)

[1599] 2. The system according to claim 1, wherein the degree of match is calculated based on the degree of match between the title and abstract of the search result and the title and abstract of the generated result.

[1600] (Claim 3)

[1601] 10. The system of claim 1, wherein the comparison and scoring of the search results and the generated results is performed using a rating engine based on the reliability of the generated results.

[1602] "Application Example 1"

[1603] (Claim 1)

[1604] means for receiving a search query;

[1605] means for obtaining search results from a search engine based on the search query;

[1606] means for transmitting the search query to an artificial intelligence model for generating the search query and obtaining the generated result;

[1607] means for comparing the search results with the generated results and calculating the degree of match;

[1608] A means for scoring the information accuracy of the search results based on the generated results;

[1609] means for re-ranking the search results based on the degree of match and the information accuracy;

[1610] means for displaying the re-ranked search results on a terminal;

[1611] A means to present optimized food delivery options using a generative AI model based on the dish name and restaurant name entered by the user; and

[1612] A system including:

[1613] (Claim 2)

[1614] 2. The system of claim 1, wherein the degree of match is calculated based on the degree of match between the title and snippet of the search result and the title and snippet of the generated result.

[1615] (Claim 3)

[1616] 2. The system of claim 1, wherein the comparison and scoring of the search results and the generated results is performed using a scoring engine based on the reliability of the generated results.

[1617] "Example 2: Combining Emotion Engines"

[1618] (Claim 1)

[1619] means for receiving a search query entered by a user;

[1620] means for obtaining search results from a search technology based on the search query;

[1621] means for transmitting the search query to an artificial intelligence model for generating the search query and obtaining the generated result;

[1622] means for comparing the search results with the generated results and calculating the degree of match;

[1623] A means for scoring the information accuracy of the search results based on the generated results;

[1624] means for re-ranking the search results based on the degree of match and the information accuracy;

[1625] means for recognizing and analyzing user emotions;

[1626] means for filtering and adjusting the re-ranked search results based on the sentiment;

[1627] means for displaying the re-ranked search results on a terminal;

[1628] A system including:

[1629] (Claim 2)

[1630] 2. The system of claim 1, wherein the degree of match is calculated based on the degree of match between the title and excerpt of the search result and the title and excerpt of the generated result.

[1631] (Claim 3)

[1632] The system of claim 1 , wherein the comparing and scoring of the search results and the generated results is performed using a scoring engine based on the reliability of the generated results.

[1633] "Application example 2 when combining emotion engines"

[1634] (Claim 1)

[1635] means for receiving a search query;

[1636] means for obtaining search results from a search engine based on the search query;

[1637] means for transmitting the search query to an artificial intelligence model for generating the search query and obtaining the generated result;

[1638] means for comparing the search results with the generated results and calculating the degree of match;

[1639] A means for scoring the information accuracy of the search results based on the generated results;

[1640] means for re-ranking the search results based on the degree of match and the information accuracy;

[1641] means for displaying the re-ranked search results on a terminal;

[1642] means for analyzing user emotions;

[1643] means for filtering or adjusting the display order of search results based on the emotion information;

[1644] A system including:

[1645] (Claim 2)

[1646] 2. The system of claim 1, wherein the degree of match is calculated based on the degree of match between the title and snippet of the search result and the title and snippet of the generated result.

[1647] (Claim 3)

[1648] 2. The system of claim 1, wherein the comparison and scoring of the search results and the generated results is performed using a scoring engine based on the reliability of the generated results. [Explanation of symbols]

[1649] 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 a search query; means for obtaining search results from a search engine based on the search query; means for transmitting the search query to an artificial intelligence model for generating the search query and obtaining the generated result; means for comparing the search results with the generated results and calculating the degree of match; A means for scoring the information accuracy of the search results based on the generated results; means for re-ranking the search results based on the degree of match and the information accuracy; means for displaying the re-ranked search results on a terminal; A system including:

2. The system of claim 1 , wherein the degree of match is calculated based on the degree of match between the title and snippet of the search result and the title and snippet of the generated result.

3. The system of claim 1 , wherein the comparison and scoring of the search results and the generated results is performed using a scoring engine based on the reliability of the generated results.

Images