System

Abstract

A system is provided.SOLUTION: A system, comprising: means for receiving an ambiguous query input by a user; means for analyzing the ambiguous query and generating an appropriate answer; means for integrating the generated answer and related information to generate a final AI result; and means for sending the final search result to the user's device for display to the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Providing appropriate search results for ambiguous search queries is difficult for conventional search systems, resulting in a poor user experience. Since users cannot be provided with the information they are looking for quickly and accurately, users' search behavior is inefficient. Furthermore, due to a lack of access to related information and flow, it can take a long time for users to find the information they are looking for. There is a need to solve these problems and improve the user experience. [Means for solving the problem]

[0005] This invention provides a system equipped with a generative AI means for receiving ambiguous queries entered by users, analyzing the queries, and generating appropriate answers. Specifically, the system obtains data from catalogs and knowledge bases maintained by companies and fine-tunes the generative AI means to provide highly accurate information. The system also includes a means for integrating the generated answers and related information to generate final search results. Furthermore, the system provides a means for including map information and related search paths in the search results, making it easier for users to access the desired information. This system enables accurate and efficient information provision even for ambiguous queries, improving the user experience.

[0006] An "ambiguous query" is a search query that does not clearly indicate a specific intent or purpose and has multiple interpretations.

[0007] "Generative AI means" refers to the artificial intelligence (AI) algorithms and models that analyze queries received from users and generate appropriate information.

[0008] A "catalog" refers to a database that compiles information about various items and services owned by a company.

[0009] A "knowledge base" refers to a database of information based on specialized knowledge, past data, and patterns.

[0010] "Fine tuning" refers to the process of fine-tuning AI performance and functionality to suit specific applications and conditions.

[0011] "Search Results" refers to the collection of answers and related information generated by the system based on a user-entered query.

[0012] "Map information" refers to visual location information about a specific place or area. [Brief explanation of the drawings]

[0013] [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

[0014] 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.

[0015] First, the terms used in the following description will be explained.

[0016] 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).

[0017] 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.

[0018] 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.

[0019] 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.

[0020] 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."

[0021] [First embodiment]

[0022] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0023] 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.

[0024] 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).

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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.

[0029] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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."

[0034] This invention is a system that uses a generative AI means to provide appropriate search results for ambiguous queries entered by a user. The system includes a user terminal, a server, and a communication means for linking these. It also uses a catalog and knowledge base maintained by a company to fine-tune the generative AI means.

[0035] Below, the processing flow of the program is shown, and a specific example of operation based on this flow is explained.

[0036] Receiving and parsing queries

[0037] 1. The user enters the search query "Kioicho" into the device and clicks the search button.

[0038] 2. The device sends the query "Kioicho" to the server.

[0039] 3. The server analyzes the received query and identifies its ambiguities.

[0040] Generating appropriate answers using generative AI methods

[0041] 1. The server retrieves relevant data from the company's catalog or knowledge base, such as information about current events in Kioicho or recommended tourist spots.

[0042] 2. The server's AI generator generates optimal answers based on this data. These answers include a list of events currently being held in Kioicho and descriptions of recommended tourist spots.

[0043] Search result consolidation and submission

[0044] 1. In addition to the generated answer, the server integrates map links to restaurants around Kioicho and related map information.

[0045] 2. The server generates the final search result data and sends it to the terminal.

[0046] Displaying search results

[0047] 1. The device displays the received search results to the user. Specifically, it displays information such as a list of events being held in Kioicho, a list of tourist spots, and a map link to nearby restaurants.

[0048] 2. The user checks the displayed information, selects events or spots that interest them, and clicks on map links if necessary, which allows the user to obtain information about nearby restaurants.

[0049] Specific examples

[0050] For example, if a user enters the query "Kioicho," the following search results will be displayed:

[0051] List of current events in Kioicho

[0052] "2023 Spring Festival"

[0053] Weekend live music

[0054] "Art Gallery Exhibition"

[0055] Recommended tourist spots

[0056] "Kioicho Garden"

[0057] "History Museum"

[0058] "Shopping mall"

[0059] Map link to nearby restaurants

[0060] "Restaurant A"

[0061] "Cafe B"

[0062] "Bar C"

[0063] In this way, users can quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. This system is expected to significantly improve user experience.

[0064] The processing flow will be explained below.

[0065] Step 1:

[0066] The user enters the query "Kioicho" into the search box on the device and presses the search button.

[0067] Step 2:

[0068] The terminal sends the query "Kioicho" entered by the user to the server.

[0069] Step 3:

[0070] The server analyzes the received query, determines that it is an ambiguous query, and determines that it needs to use generative AI methods to generate an appropriate answer.

[0071] Step 4:

[0072] The server accesses the company's catalog database to retrieve the latest event information about Kioicho and related tourist spots, and also retrieves related information from the knowledge base.

[0073] Step 5:

[0074] The server's AI generation method generates optimal answers based on the acquired data, such as a list of events currently being held in Kioicho or descriptions of recommended tourist spots.

[0075] Step 6:

[0076] The server will then integrate additional information into the generated answer, specifically including map links to restaurants around Kioicho and related map information in the search results.

[0077] Step 7:

[0078] The server generates the final search result data and sends it to the user's device. The search result data includes the generated answer, an event list, a list of tourist attractions, and a map link.

[0079] Step 8:

[0080] The device analyzes the search results it receives and displays them in an easy-to-understand format for the user, such as a list of events currently being held in Kioicho, a list of tourist spots, and map links to nearby restaurants.

[0081] Step 9:

[0082] Users can check the search results and select events or spots they are interested in. If necessary, they can click on a map link to get more detailed information.

[0083] Example 1

[0084] 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."

[0085] Current search systems have a problem in that they cannot provide sufficiently appropriate results for ambiguous queries entered by users. In particular, when users search without a specific intent, it is difficult to efficiently retrieve relevant information. Furthermore, search results often do not integrate map information or related links, which does not improve the user experience. Therefore, there is a need for a system that can quickly and accurately find the desired information even when users enter ambiguous queries.

[0086] 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.

[0087] In this invention, the server includes means for receiving an ambiguous query entered by a user, means for analyzing the ambiguous query and identifying its ambiguity, means for retrieving related data from a database maintained by the company, means for generating an optimal answer using a generative AI model, means for integrating the generated answer and related map information to generate final search results, and means for transmitting the final search results to the user's terminal and displaying them to the user, thereby enabling the user to quickly and accurately find desired information even for an ambiguous query.

[0088] An "ambiguous query" is a general or vague search term entered by a user without specifying a specific intent or detailed information.

[0089] A "generative AI model" refers to an artificial intelligence that is trained on large datasets and is capable of natural language generation and analysis.

[0090] A "Company-maintained database" is a collection of various sources owned and controlled by the Company that contain information relevant to a particular query.

[0091] A "best answer" refers to the response that provides the most relevant and valuable information to the query entered by the user.

[0092] "Map Information" means information showing locations and routes for a particular geographic area, which may be obtained from external services such as APIs.

[0093] An "HTTP POST request" is one of the communication protocols on the Internet that allows a terminal to send data to a server, and refers to a request that includes the user's query data.

[0094] "Natural language processing (NLP) algorithms" refers to the technologies and methods that enable computers to understand, analyze, and generate human language.

[0095] An "SQL database" is a type of information system for managing and manipulating data using a structured query language.

[0096] "Google Maps API" is an application programming interface for the map information service provided by Google, and is used to obtain specific geographic information.

[0097] "Fine-tuning" refers to additional training to tailor a generative AI model to suit a specific task and data.

[0098] This invention is a system that uses a generative AI means to provide appropriate search results for ambiguous queries entered by a user. The system includes a user terminal, a server, and a communication means for linking these. It also uses a catalog and knowledge base maintained by a company to fine-tune the generative AI means.

[0099] Receiving and parsing queries

[0100] A user enters the query "Kioicho" into a device and clicks the search button. The device sends this query "Kioicho" to the server. The server analyzes the received query and identifies its ambiguity. The analysis uses natural language processing (NLP) algorithms to tokenize the query and identify grammatical ambiguity.

[0101] Generating appropriate answers using generative AI methods

[0102] The server retrieves relevant data from a database maintained by the company. This data includes information on current events in Kioicho and data on recommended tourist spots. An SQL database is used for this retrieval. The server's generative AI model (e.g., GPT-4) then generates the optimal answer based on this data. The generated answer includes a list of current events in Kioicho and a description of recommended tourist spots.

[0103] Search result consolidation and submission

[0104] The server integrates the generated answers with map links to restaurants in the Kioicho area and other related map information, including map information obtained from external services such as Google Maps API, to generate the final search result data, which is then sent to the device.

[0105] Displaying search results

[0106] The device displays the received search results to the user. Specifically, it displays information such as a list of events being held in Kioicho, a list of tourist spots, and map links to nearby restaurants. The user can check the displayed information, select events or spots that interest them, and click on map links as needed. This allows the user to obtain information about nearby restaurants.

[0107] Specific examples

[0108] For example, if a user enters the query "Kioicho," the following search results will be displayed:

[0109] List of current events in Kioicho

[0110] "2023 Spring Festival"

[0111] Weekend live music

[0112] "Art Gallery Exhibition"

[0113] Recommended tourist spots

[0114] "Kioicho Garden"

[0115] "History Museum"

[0116] "Shopping mall"

[0117] Map link to nearby restaurants

[0118] "Restaurant A"

[0119] "Cafe B"

[0120] "Bar C"

[0121] Example prompts for generative AI models

[0122] Query: "Kioicho"

[0123] Task for the generative AI: "In response to a vague user query such as 'Kioicho,' generate an answer that includes a list of current events in Kioicho, recommended tourist spots, and map links to nearby restaurants. Relevant information should be obtained from company catalogs and knowledge bases. Map information should be provided using the Google Maps API."

[0124] This invention enables users to quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. In particular, it is expected that the user experience will be significantly improved by providing search results that integrate related map information and event information.

[0125] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0126] Processing Steps

[0127] Step 1:

[0128] Entering a query

[0129] The user enters the query "Kioicho" into the terminal and clicks the search button.

[0130] Specific actions: A user types "Kioicho" into the search bar of a web browser and presses the enter key, or performs a similar operation using a dedicated search app.

[0131] Input: The vague query "Kioicho" entered by the user.

[0132] Output: User input is saved to the terminal.

[0133] Step 2:

[0134] Submitting a query

[0135] The terminal sends the query "Kioicho" entered by the user to the server.

[0136] Specific operation: The device sends an HTTP POST request to the server and includes the query data.

[0137] Input: The device saved query "Kioicho".

[0138] Output: The query is sent to the server.

[0139] Step 3:

[0140] Parsing a query

[0141] The server analyzes the received query and identifies any ambiguities.

[0142] What happens: The server uses a natural language processing (NLP) library (e.g., spaCy or BERT) to tokenize the query and identify grammatical ambiguities.

[0143] Input: The query "Kioicho" received by the server.

[0144] Output: The analysis results in identifying ambiguities.

[0145] Step 4:

[0146] Retrieving related data

[0147] The server retrieves the relevant data from a database maintained by the company.

[0148] Specific operation: The server executes a query against the SQL database to retrieve relevant information, such as "information about current events in Kioicho" or "data about recommended tourist spots."

[0149] Input: A query based on the analysis results.

[0150] Output: Related data obtained (event information, tourist spot information, etc.).

[0151] Step 5:

[0152] Generate answers

[0153] The server's generative AI means (e.g., GPT-4) generates the optimal answer based on the acquired data.

[0154] What it does: The generative AI model takes data as input and creates an appropriate response based on the prompt, for example, listing "All events currently happening in Kioicho."

[0155] Input: The retrieved data and the prompt statement.

[0156] Output: Generated answers (event list, tourist attraction description, etc.).

[0157] Step 6:

[0158] Integration of results

[0159] In addition to the generated answers, the server integrates map links to restaurants around Kioicho and related map information.

[0160] Specific operation: The server calls the Google Maps API to obtain map information for restaurants and tourist spots related to "Kioicho" and includes it in the generated answer.

[0161] Input: Generated answers and map information obtained from the Google Maps API.

[0162] Output: The final consolidated search results.

[0163] Step 7:

[0164] Sending the results

[0165] The server generates the final search result data and transmits it to the terminal.

[0166] Specific operation: The server compiles the generated answer and map information and sends it to the terminal as an HTTP response.

[0167] Input: Consolidated search results data.

[0168] Output: Search result data sent to the device.

[0169] Step 8:

[0170] Displaying the results

[0171] The terminal displays the received search results to the user.

[0172] What it does: The device uses HTML and CSS to render the content it receives as a web page and displays it to the user.

[0173] Input: The received search result data.

[0174] Output: Search results displayed to the user.

[0175] Step 9:

[0176] Verify the information

[0177] The user checks the displayed information, selects events or spots of interest, and clicks on map links if necessary.

[0178] What Happens: A user scrolls through the search results and clicks a link that interests them (e.g., event details or a map). The browser opens the link and displays additional information.

[0179] Input: The search results shown to the user.

[0180] Output: Additional information displayed based on the link the user selects.

[0181] (Application example 1)

[0182] 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."

[0183] In physical stores, it is difficult to provide quick and accurate information in response to vague queries entered by users. This can result in customers not being able to obtain the information they need accurately, which can lead to lower satisfaction. There is also a need to provide a means for customers visiting physical stores to instantly access various information within the store.

[0184] 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.

[0185] In this invention, the server includes means for receiving an ambiguous query entered by a user, a generation AI means for analyzing the ambiguous query and generating an appropriate answer, a means for integrating the generated answer and related information to generate a final search result, a means for transmitting the search result to the user's terminal and displaying it to the user, and a means for allowing the user to obtain search information via hardware located in a physical store. This makes it possible to provide information in response to a user's ambiguous query quickly and accurately even in a physical store, and is expected to improve customer satisfaction.

[0186] An "ambiguous query" is a search request or question entered by a user without a clear, specific intent.

[0187] "Generative AI Means" means a processing means that employs artificial intelligence techniques to generate appropriate responses to received queries.

[0188] A "catalog" refers to a database that lists product information and service details held by a company.

[0189] A "knowledge base" refers to a database that systematically organizes a company's specialized knowledge and past data.

[0190] "Hardware" refers to any physical device or equipment used to process information, including smartphones, smart glasses, and tablet devices.

[0191] "User device" refers to a device that a user directly operates to input and receive information, including smartphones and tablet devices.

[0192] "Means for generating search results" refers to the processing means that aggregates and organizes the final search results based on the answers obtained by the AI ​​generation means in response to ambiguous queries, and provides them to the user.

[0193] "Map information" refers to data containing geographical information about a specific area or location, such as the location of a physical store or nearby facilities.

[0194] "Means for obtaining search information" refers to the technological means that enable users to effectively search for and obtain the information they need in a physical store, including tablet devices and information kiosks installed in the store.

[0195] This invention provides a system for providing quick and accurate information in response to ambiguous queries entered by users in a physical store. The system is implemented using a user terminal, a server, and hardware installed in the physical store.

[0196] System Configuration

[0197] User devices, including smartphones and tablets, where users enter ambiguous queries.

[0198] Server: The server mainly performs the following processes:

[0199] 1. Receiving a query: Receive a query sent from a user terminal.

[0200] 2. Query Analysis: Analyze the ambiguity of the received query and generate an appropriate answer.

[0201] 3. Data Acquisition: Obtain relevant data from catalogs and knowledge bases maintained by the company.

[0202] 4. Generative AI method: Based on this data, the optimal answer is generated using generative AI method.

[0203] 5. Integration and transmission of search results: The generated answers and related information are integrated to generate the final search results and transmit them to the user terminal.

[0204] Processing details

[0205] The server uses OpenAI's GPT-3 model as its generative AI method, which allows it to generate highly accurate answers even for ambiguous queries. Specifically, if a user enters the query "What are the recommended sale items?", the following prompt sentence is input to the generative AI model:

[0206] What sale items do you recommend?

[0207] Catalog: [Sales and product information]

[0208] Knowledge Base: [Store recommendations, customer reviews, sale dates, etc.]

[0209] Use on devices

[0210] Hardware installed in physical stores: Information kiosks and tablet devices are installed in stores, and users can use them to search for information. For example, if a user types "What are some recommended restaurants nearby?" into a tablet device, the server retrieves relevant data and generates the optimal answer using generative AI methods. The generated answer includes a list of nearby restaurants and map information.

[0211] Hardware and Software Use

[0212] The server is implemented using the Python programming language and operates the generative AI model through the OpenAI API. Communication between the user device and the server uses the HTTPS protocol. The hardware placed in the physical store includes tablet devices and information kiosks, which communicate with the server to obtain and display information in real time.

[0213] This system is expected to improve customer satisfaction by allowing users to quickly and accurately obtain the information they need in physical stores.

[0214] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0215] Step 1:

[0216] Receiving a query

[0217] A user enters a vague query on a smartphone or tablet in a physical store, for example, "What are some recommended sale items?"

[0218] Input: A vague query entered by the user

[0219] Output: Query data for the device to send to the server

[0220] Step 2:

[0221] Submitting a query

[0222] The device sends the entered fuzzy query to the server over the HTTPS protocol.

[0223] Input: An ambiguous query typed into the terminal

[0224] Output: Query data received by the server

[0225] Step 3:

[0226] Parsing a query

[0227] The server analyzes the ambiguous query and identifies the ambiguity, for example, what is a "sale item."

[0228] Input: The ambiguous query received by the server

[0229] Output: Parsed query information (e.g., keyword "sale items")

[0230] Step 4:

[0231] Data Acquisition

[0232] The server retrieves the necessary data from the company's catalog and knowledge base.

[0233] Input: Parsed query information

[0234] Output: Catalog and Knowledge Base data

[0235] Step 5:

[0236] Generate prompt statement

[0237] The server generates prompt sentences to input into the generative AI model based on the acquired data.

[0238] Input: Catalog data and knowledge base data

[0239] Output: A prompt to be input to the generative AI model

[0240] Step 6:

[0241] Answer generation using generative AI methods

[0242] The server uses generative AI methods (e.g., OpenAI GPT-3 model) to generate the best answer based on the prompt.

[0243] Input: prompt statement

[0244] Output: Generated answer (e.g. "This week's featured sale items are...")

[0245] Step 7:

[0246] Search result integration

[0247] The server integrates the generated answers with related information (e.g., maps and links) to generate the final search results.

[0248] Input: Generated answers and related information

[0249] Output: Consolidated search results data

[0250] Step 8:

[0251] Submit search results

[0252] The server sends the final search results to the user's terminal.

[0253] Input: Integrated search result data

[0254] Output: Search result data received by the device

[0255] Step 9:

[0256] Displaying search results

[0257] The device then displays the search results to the user, such as a list of sale items or a map link.

[0258] Input: Search result data received by the device

[0259] Output: Search results displayed so that users can see them visually

[0260] These specific processing steps enable users to quickly and accurately obtain the information they need in a physical store.

[0261] 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.

[0262] This invention is a system that uses a generative AI means and an emotion engine to provide appropriate search results for ambiguous queries entered by users. The system includes a user terminal, a server, an emotion engine, and a communication means for linking these. It also uses a catalog and knowledge base maintained by a company to fine-tune the generative AI means.

[0263] Below, the processing flow of the program is shown, and a specific example of operation based on this flow is explained.

[0264] Receiving and parsing queries

[0265] 1. The user enters the query "Kioicho" into the search box on the device and presses the search button.

[0266] 2. The device sends the query "Kioicho" to the server.

[0267] 3. The server analyzes the received query and identifies its ambiguities.

[0268] Generative AI methods and emotion engines generate appropriate answers

[0269] 1. The server accesses the company's catalog database to retrieve the latest event information about Kioicho and related tourist spots. It also retrieves related information from the knowledge base.

[0270] 2. The server's AI generates optimal answers based on the acquired data. The generated answers include a list of events currently being held in Kioicho and descriptions of recommended tourist spots.

[0271] 3. The emotion engine recognizes and analyzes the user's emotions based on their input and past behavioral data. For example, it can determine emotions such as excitement, joy, or sadness based on the user's typing speed and choice of words.

[0272] 4. The server optimizes the response provided by the generative AI means based on the emotional information recognized by the emotion engine. For example, if the user is feeling stressed, the server may recommend relaxing tourist spots.

[0273] Search result consolidation and submission

[0274] 1. In addition to the generated answer, the server integrates map links to restaurants around Kioicho and related map information.

[0275] 2. The server adjusts the display format of search results according to the user's emotions recognized by the emotion engine. For example, if the user feels like relaxing, the server displays the results in a calming interface.

[0276] 3. The server generates the final search result data and sends it to the user's device. The search result data includes the generated answer, event list, tourist attraction list, and map link.

[0277] Displaying search results

[0278] 1. The device displays the received search results to the user. Specifically, it displays information such as a list of events being held in Kioicho, a list of tourist spots, and a map link to nearby restaurants.

[0279] 2. The user reviews the search results and selects the events or spots they are interested in. If necessary, they can click on the map link to obtain more detailed information.

[0280] Specific examples

[0281] For example, if a user enters the query "Kioicho," the following search results will be displayed:

[0282] List of current events in Kioicho

[0283] "2023 Spring Festival"

[0284] Weekend live music

[0285] "Art Gallery Exhibition"

[0286] Recommended tourist spots

[0287] "Kioicho Garden"

[0288] "History Museum"

[0289] "Shopping mall"

[0290] Map link to nearby restaurants

[0291] "Restaurant A"

[0292] "Cafe B"

[0293] "Bar C"

[0294] Furthermore, if the system detects that the user is feeling stressed, it will prioritize tourist spots that are relaxing (e.g., Kioicho Garden) and display the search results in a calming color scheme, providing a user-friendly UI.

[0295] In this way, users can quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. This system is expected to significantly improve user experience.

[0296] The processing flow will be explained below.

[0297] Step 1:

[0298] The user enters the query "Kioicho" into the search box on the device and presses the search button.

[0299] Step 2:

[0300] The terminal sends the query "Kioicho" entered by the user to the server.

[0301] Step 3:

[0302] The server analyzes the received query, determines that it is an ambiguous query, and determines that it needs to use generative AI methods to generate an appropriate answer.

[0303] Step 4:

[0304] The server accesses the company's catalog database to retrieve the latest event information about Kioicho and related tourist spots, and also retrieves related information from the knowledge base.

[0305] Step 5:

[0306] The server's AI generator generates optimal answers based on the acquired data. These answers include a list of events currently being held in Kioicho and descriptions of recommended tourist spots.

[0307] Step 6:

[0308] The emotion engine recognizes and analyzes the user's emotions at that time based on their input, past behavioral data, facial expressions, tone of voice, etc. For example, it can determine emotions such as excitement, joy, or sadness based on the user's input speed and choice of words.

[0309] Step 7:

[0310] The server optimizes the response provided by the generative AI means based on the emotional information recognized by the emotion engine. For example, if the user is feeling stressed, the server will adjust the response to recommend relaxing tourist spots.

[0311] Step 8:

[0312] In addition to the generated answers, the server integrates map links to restaurants around Kioicho and related map information.

[0313] Step 9:

[0314] The server adjusts the display format of search results according to the user's emotions recognized by the emotion engine. For example, if the user feels like relaxing, the server displays the results in a calming interface.

[0315] Step 10:

[0316] The server generates the final search result data and sends it to the user's device. The search result data includes the generated answer, an event list, a list of tourist attractions, and a map link.

[0317] Step 11:

[0318] The device then displays the received search results to the user, such as a list of events currently being held in Kioicho, a list of tourist spots, and map links to nearby restaurants.

[0319] Step 12:

[0320] Users can check the search results and select events or spots they are interested in. If necessary, they can click on a map link to get more detailed information.

[0321] Example 2

[0322] 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."

[0323] Conventional search systems have difficulty providing appropriate answers to ambiguous queries entered by users, and they are unable to provide appropriate search results that reflect the user's emotions, resulting in a poor user experience.

[0324] 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.

[0325] In this invention, the server includes means for receiving an ambiguous query entered by a user, generation AI means for analyzing the ambiguous query and generating an appropriate answer, means for integrating the generated answer and related information to generate final search results, an emotion engine for recognizing and analyzing emotions from the user's input and past behavioral data, and means for optimizing the answer by the generation AI means based on the emotion information recognized by the emotion engine. This allows the user to quickly obtain appropriate search results even for an ambiguous query, and further provides an optimal answer according to the user's emotions, thereby improving the user experience.

[0326] A "query" is a word or phrase that a user enters to search for information.

[0327] A "generative AI means" is an artificial intelligence technology for analyzing a received query and generating an appropriate response to that query.

[0328] The "emotion engine" is a function that recognizes and analyzes emotions from user input and past behavioral data.

[0329] A "database" is a collection of specific information that is systematically organized, stored, and can be accessed as needed.

[0330] A "knowledge base" is a database that systematically organizes and accumulates specific specialized knowledge and information.

[0331] "Cross-reference" is a function or method that allows related information to be mutually referenced.

[0332] The "integration means" is a function for compiling the generated answers and related information into a single result and providing it to the user.

[0333] "Optimization" is the process of adjusting or improving to obtain the best possible condition or result for a particular purpose or condition.

[0334] "Search results" are a collection of answers and related information provided in response to a user-entered query.

[0335] This invention is a system that uses a generative AI means and an emotion engine to provide appropriate search results for ambiguous queries entered by users. The system includes a user terminal, a server, an emotion engine, and a communication means for linking these. Furthermore, the system uses a database and knowledge base maintained by a company to fine-tune the generative AI means.

[0336] System Overview

[0337] User terminal

[0338] A user terminal is a device through which a user enters a query and receives and displays search results. Specific hardware examples include PCs, smartphones, and tablets. Users enter queries and view results through this terminal.

[0339] server

[0340] The server is responsible for parsing queries, generating answers using generative AI means, recognizing and analyzing emotions using an emotion engine, and integrating and transmitting search results. The server is equipped with a high-performance processor and a large amount of memory, allowing for real-time data processing.

[0341] Generation AI means

[0342] Generative AI is used to analyze queries and generate appropriate answers. Specifically, it uses machine learning algorithms to analyze queries and generate optimal answers based on relevant information. This generative AI is fine-tuned using the company's database and knowledge base.

[0343] Emotion Engine

[0344] The emotion engine is an engine that recognizes and analyzes emotions from user input and past behavioral data. Specifically, it analyzes data such as input speed, word choice, and query frequency to determine whether the user is feeling excitement, joy, sadness, stress, or other emotions.

[0345] Specific examples

[0346] An example of the behavior when a user enters the query "Kioicho" into the search box of a terminal will be described.

[0347] 1. Receiving a query: The user enters the query "Kioicho" in the search box of the device and presses the search button. The device sends this query to the server.

[0348] 2. Query Analysis: The server analyzes the received query and identifies its ambiguity. It determines that "Kioicho" is a place name and is associated with a wide range of data, including information about nearby tourist attractions and events.

[0349] 3. Data Acquisition: The server acquires the latest event information and tourist spot data about Kioicho from the company's database and knowledge base.

[0350] 4. Answer generation: Based on the data acquired by the generation AI means, the optimal answer is generated, including descriptions of tourist spots such as "Kioicho Garden" and "Historical Museum," and event information such as "2023 Spring Festival" and "Weekend Music Live."

[0351] 5. Emotion recognition and analysis: The emotion engine recognizes that the user is likely feeling stressed based on their typing speed and choice of words.

[0352] 6. Answer optimization: The server optimizes the answers based on the emotional information recognized by the emotion engine, so that relaxing spots (e.g., "Kioicho Garden") are given priority.

[0353] 7. Integration and transmission of search results: The server integrates the generated answers with map links to restaurants around Kioicho and related map information to generate the final search results and transmit them to the user's device.

[0354] 8. Displaying search results: The device displays the received search results to the user. Specifically, it displays a list of events being held in Kioicho, Kioicho Garden, and a map link to nearby restaurants. The user can check this information and click to view more detailed information as needed.

[0355] In this way, users can quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. Utilizing an emotion engine is expected to significantly improve the quality of the user experience.

[0356] Prompt Sentence Examples

[0357] As an example of a user entering a query, consider the following prompt:

[0358] "Please tell me about historical tourist spots in Kioicho."

[0359] "What events are happening in Kioicho?"

[0360] "Please tell me some recommended restaurants around Kioicho."

[0361] Searches are performed based on these prompts, and generative AI methods and an emotion engine work together to provide optimal search results.

[0362] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0363] Step 1:

[0364] Receiving a query

[0365] The user enters the query "Kioicho" into the search box on the device and presses the search button.

[0366] Input: User-entered query "Kioicho"

[0367] Output: Data sent from the device to the server for the query "Kioicho"

[0368] Specific operation: When the search button on the terminal interface is clicked, query data is constructed and sent to the server.

[0369] Step 2:

[0370] Parsing a query

[0371] The server analyzes the received query and identifies any ambiguities.

[0372] Input: Query "Kioicho" sent from the terminal

[0373] Output: Information about the ambiguity of the query (e.g., it is a place name and may be related to tourist attractions or event information)

[0374] What happens: The server uses natural language processing algorithms to analyze the meaning of the query and identify any ambiguities.

[0375] Step 3:

[0376] Data Acquisition

[0377] The server retrieves the latest event information and tourist spot data about Kioicho from the company's database and knowledge base.

[0378] Input: Query ambiguity information and related database lookup request

[0379] Output: Related event information and tourist spot data (e.g., "2023 Spring Festival," "Kioicho Garden," etc.)

[0380] Specific operation: The server calls the database API, obtains the information corresponding to the query, and stores it in the cache.

[0381] Step 4:

[0382] Generate answers

[0383] The generation AI means generates the optimal answer based on the data acquired.

[0384] Input: Event information and tourist spot data obtained from the database

[0385] Output: A specific answer to be provided to the user (e.g., "Kioicho Garden is a beautiful garden and a great place to relax," "There is live music on weekends," etc.)

[0386] Specific operation: The generative AI means converts the acquired data using a natural language generation algorithm to make it easier for humans to understand.

[0387] Step 5:

[0388] Emotion Recognition and Analysis

[0389] The emotion engine recognizes and analyzes the user's emotions at that time based on their input speed and choice of words.

[0390] Input: User typing speed, words used, and past behavior data

[0391] Output: Information about the user's emotional state (e.g., feeling stressed)

[0392] How it works: The emotion engine uses machine learning algorithms to determine emotions from input data.

[0393] Step 6:

[0394] Optimizing answers

[0395] The server optimizes the answer provided by the generative AI means based on the emotional information recognized by the emotion engine.

[0396] Input: Generated answers and information about the user's emotional state

[0397] Output: Optimized answers (e.g., prioritize relaxation spots for stressed users)

[0398] What happens: The server re-evaluates the answer and adjusts it accordingly.

[0399] Step 7:

[0400] Search result consolidation and submission

[0401] The server then integrates the generated answers with map links to restaurants around Kioicho and other related map information to generate the final search results, which are then sent to the user's device.

[0402] Input: Optimized answer and additional information (e.g. restaurant map link, map information)

[0403] Output: Final search result data

[0404] Specific operation: The server compiles information from various data sources, constructs the final search results, and sends them to the device.

[0405] Step 8:

[0406] Displaying search results

[0407] The terminal displays the received search results to the user.

[0408] Input: Final search result data sent from the server

[0409] Output: Search results displayed on the screen (e.g., "List of events being held in Kioicho," "Kioicho Garden," "Map links to nearby restaurants")

[0410] What happens: The device analyzes the received data and displays it appropriately in the user interface. The user can click on the information of interest to view further details.

[0411] (Application example 2)

[0412] 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."

[0413] Conventional search systems have difficulty providing satisfactory search results when users enter ambiguous queries. Providing optimal search results is particularly challenging when the information desired varies depending on the user's emotions. Food delivery services also face challenges in recommending appropriate restaurants and menus based on ambiguous queries and user emotions.

[0414] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving an ambiguous query entered by a user; a generation AI means for analyzing the ambiguous query and generating an appropriate answer; an emotion engine means for analyzing the user's emotions and optimizing the generated answer and related information; a means for integrating the generated answer and related information to generate final search results; and a means for transmitting the search results to the user's terminal and displaying them to the user. This makes it possible to provide optimal search results that take the user's emotions into consideration, even when the user enters an ambiguous query. Furthermore, in a food delivery service, optimal restaurants and menus can be recommended based on the ambiguous query, thereby improving user satisfaction.

[0415] An "ambiguous query" is an unclear search term that a user enters without providing any specific information.

[0416] "Generative AI methods" refers to artificial intelligence techniques for analyzing ambiguous queries and generating appropriate answers.

[0417] The "emotion engine" is an engine that analyzes emotions from user input and behavioral data and optimizes the generated information.

[0418] "User terminal" refers to the device used by the user to perform a search, including smartphones and personal computers.

[0419] A "corporate catalog" is a database that lists the products and services offered by a company.

[0420] A "knowledge base" is a database that systematically compiles the knowledge and information held by a company.

[0421] "Search Results" refers to answers and related information generated based on a user-entered query.

[0422] "Map information" is data that indicates geographical location information included in the search results.

[0423] "Route information" is data that provides guidance for a user to reach a particular location.

[0424] This invention is a system that provides optimal search results for ambiguous queries entered by users using generative AI means and an emotion engine. This system is embodied as an application called "FoodGenie" that is specialized for food delivery services.

[0425] Program processing

[0426] 1. Receiving a query:

[0427] When a user enters an ambiguous query (e.g., "pizza") on a smartphone app, the device sends this query to a server.

[0428] 2. Emotion analysis:

[0429] The server sends the query to the emotion engine, which analyzes the user's emotional state based on the user's input speed and word choice.

[0430] 3. Answer generation by generative AI:

[0431] The server accesses the company's catalog or knowledge base to retrieve the necessary data, and the generative AI method generates the appropriate answer based on this.

[0432] 4. Emotion-based optimization:

[0433] The generated answers are optimized based on the analysis results from the emotion engine. For example, if the user is feeling stressed, it will recommend a "relaxing family pizza set."

[0434] 5. Search Results Integration:

[0435] The server integrates the generated answers, along with any associated map or route information, to produce the final search results.

[0436] 6. Submitting and Displaying Search Results:

[0437] The server sends the final search results to the user's device and displays them to the user, who then displays the results in the app's interface.

[0438] Hardware and software used

[0439] Device: Smartphone (iOS, Android)

[0440] Emotion Engine API: Emotion-API Platform

[0441] Generate AI model: Custom AI Model API (e.g. OpenAI, GPT-4)

[0442] Server: Cloud server (AWS, Google Cloud)

[0443] Specific examples

[0444] For example, if a user types "pizza" into their smartphone, the emotion engine will detect "stress." The generative AI will prioritize recommendations for restaurants that offer "relaxing family pizza sets."

[0445] Example prompt for a generative AI model:

[0446] plaintext

[0447] A user types "pizza." The user's emotion is "stress." Can you recommend a restaurant and menu that will help them relax and relieve stress?

[0448] This system can provide optimal search results that take emotions into account, even when a user enters an ambiguous query.Furthermore, in food delivery services, it can recommend optimal restaurants and menus based on ambiguous queries, thereby improving user satisfaction.

[0449] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0450] Step 1:

[0451] A user enters a vague query into the "FoodGenie" app on their smartphone and presses the search button. The entered query is a non-specific phrase such as "pizza." Once the input is received, the device sends the query information to the server. The input data includes the query text and the user ID.

[0452] Step 2:

[0453] The server sends a request to the emotion engine API to analyze the received query. The emotion engine takes the query text and user ID as input and analyzes the user's emotional state. For example, it determines the user's emotion (e.g., stress or fatigue) from the query input speed and the words used. The emotion information is returned to the server.

[0454] Step 3:

[0455] The server accesses the company's catalog and knowledge base to gather data related to the query. For example, the server retrieves multiple restaurant and menu data related to "pizza." The results collected from the database include restaurant names, menus, ratings, location information, etc.

[0456] Step 4:

[0457] The server uses a generative AI to analyze the target data and generate the optimal answer. At this time, the server uses the collected restaurant data and the user's emotional information obtained from the emotion engine as input. For example, if it is determined that the user is feeling stressed, the generative AI will suggest a "relaxing family pizza set to relieve stress." The generated answer data includes information such as the restaurant name, recommended menu items, and ratings.

[0458] Step 5:

[0459] The server further integrates the generated answers and refines the final search results. Specifically, it adds map information and route information for the restaurant. It also optimizes the interface based on the analysis results of the emotion engine (for example, displaying calmer colors for users with high stress levels). The final search result data is generated.

[0460] Step 6:

[0461] The server then sends the final search results to the user's device, which then displays them on the app's interface, including the name of the restaurant the user is looking for, map information, and recommended menu items.

[0462] Step 7:

[0463] Users can review the search results displayed in the app and select the restaurant or menu they are interested in. This allows users to easily obtain information and directions to specific restaurants.

[0464] 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.

[0465] 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.

[0466] 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.

[0467] [Second embodiment]

[0468] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0469] 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.

[0470] 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).

[0471] 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.

[0472] 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.

[0473] 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).

[0474] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0475] 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.

[0476] 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.

[0477] 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.

[0478] 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.

[0479] 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."

[0480] This invention is a system that uses a generative AI means to provide appropriate search results for ambiguous queries entered by a user. The system includes a user terminal, a server, and a communication means for linking these. It also uses a catalog and knowledge base maintained by a company to fine-tune the generative AI means.

[0481] Below, the processing flow of the program is shown, and a specific example of operation based on this flow is explained.

[0482] Receiving and parsing queries

[0483] 1. The user enters the search query "Kioicho" into the device and clicks the search button.

[0484] 2. The device sends the query "Kioicho" to the server.

[0485] 3. The server analyzes the received query and identifies its ambiguities.

[0486] Generating appropriate answers using generative AI methods

[0487] 1. The server retrieves relevant data from the company's catalog or knowledge base, such as information about current events in Kioicho or recommended tourist spots.

[0488] 2. The server's AI generator generates optimal answers based on this data. These answers include a list of events currently being held in Kioicho and descriptions of recommended tourist spots.

[0489] Search result consolidation and submission

[0490] 1. In addition to the generated answer, the server integrates map links to restaurants around Kioicho and related map information.

[0491] 2. The server generates the final search result data and sends it to the terminal.

[0492] Displaying search results

[0493] 1. The device displays the received search results to the user. Specifically, it displays information such as a list of events being held in Kioicho, a list of tourist spots, and a map link to nearby restaurants.

[0494] 2. The user checks the displayed information, selects events or spots that interest them, and clicks on map links if necessary, which allows the user to obtain information about nearby restaurants.

[0495] Specific examples

[0496] For example, if a user enters the query "Kioicho," the following search results will be displayed:

[0497] List of current events in Kioicho

[0498] "2023 Spring Festival"

[0499] Weekend live music

[0500] "Art Gallery Exhibition"

[0501] Recommended tourist spots

[0502] "Kioicho Garden"

[0503] "History Museum"

[0504] "Shopping mall"

[0505] Map link to nearby restaurants

[0506] "Restaurant A"

[0507] "Cafe B"

[0508] "Bar C"

[0509] In this way, users can quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. This system is expected to significantly improve user experience.

[0510] The processing flow will be explained below.

[0511] Step 1:

[0512] The user enters the query "Kioicho" into the search box on the device and presses the search button.

[0513] Step 2:

[0514] The terminal sends the query "Kioicho" entered by the user to the server.

[0515] Step 3:

[0516] The server analyzes the received query, determines that it is an ambiguous query, and determines that it needs to use generative AI methods to generate an appropriate answer.

[0517] Step 4:

[0518] The server accesses the company's catalog database to retrieve the latest event information about Kioicho and related tourist spots, and also retrieves related information from the knowledge base.

[0519] Step 5:

[0520] The server's AI generation method generates optimal answers based on the acquired data, such as a list of events currently being held in Kioicho or descriptions of recommended tourist spots.

[0521] Step 6:

[0522] The server will then integrate additional information into the generated answer, specifically including map links to restaurants around Kioicho and related map information in the search results.

[0523] Step 7:

[0524] The server generates the final search result data and sends it to the user's device. The search result data includes the generated answer, an event list, a list of tourist attractions, and a map link.

[0525] Step 8:

[0526] The device analyzes the search results it receives and displays them in an easy-to-understand format for the user, such as a list of events currently being held in Kioicho, a list of tourist spots, and map links to nearby restaurants.

[0527] Step 9:

[0528] Users can check the search results and select events or spots they are interested in. If necessary, they can click on a map link to get more detailed information.

[0529] Example 1

[0530] 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."

[0531] Current search systems have a problem in that they cannot provide sufficiently appropriate results for ambiguous queries entered by users. In particular, when users search without a specific intent, it is difficult to efficiently retrieve relevant information. Furthermore, search results often do not integrate map information or related links, which does not improve the user experience. Therefore, there is a need for a system that can quickly and accurately find the desired information even when users enter ambiguous queries.

[0532] 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.

[0533] In this invention, the server includes means for receiving an ambiguous query entered by a user, means for analyzing the ambiguous query and identifying its ambiguity, means for retrieving related data from a database maintained by the company, means for generating an optimal answer using a generative AI model, means for integrating the generated answer and related map information to generate final search results, and means for transmitting the final search results to the user's terminal and displaying them to the user, thereby enabling the user to quickly and accurately find desired information even for an ambiguous query.

[0534] An "ambiguous query" is a general or vague search term entered by a user without specifying a specific intent or detailed information.

[0535] A "generative AI model" refers to an artificial intelligence that is trained on large datasets and is capable of natural language generation and analysis.

[0536] A "Company-maintained database" is a collection of various sources owned and controlled by the Company that contain information relevant to a particular query.

[0537] A "best answer" refers to the response that provides the most relevant and valuable information to the query entered by the user.

[0538] "Map Information" means information showing locations and routes for a particular geographic area, which may be obtained from external services such as APIs.

[0539] An "HTTP POST request" is one of the communication protocols on the Internet that allows a terminal to send data to a server, and refers to a request that includes the user's query data.

[0540] "Natural language processing (NLP) algorithms" refers to the technologies and methods that enable computers to understand, analyze, and generate human language.

[0541] An "SQL database" is a type of information system for managing and manipulating data using a structured query language.

[0542] "Google Maps API" is an application programming interface for the map information service provided by Google, and is used to obtain specific geographic information.

[0543] "Fine-tuning" refers to additional training to tailor a generative AI model to suit a specific task and data.

[0544] This invention is a system that uses a generative AI means to provide appropriate search results for ambiguous queries entered by a user. The system includes a user terminal, a server, and a communication means for linking these. It also uses a catalog and knowledge base maintained by a company to fine-tune the generative AI means.

[0545] Receiving and parsing queries

[0546] A user enters the query "Kioicho" into a device and clicks the search button. The device sends this query "Kioicho" to the server. The server analyzes the received query and identifies its ambiguity. The analysis uses natural language processing (NLP) algorithms to tokenize the query and identify grammatical ambiguity.

[0547] Generating appropriate answers using generative AI methods

[0548] The server retrieves relevant data from a database maintained by the company. This data includes information on current events in Kioicho and data on recommended tourist spots. An SQL database is used for this retrieval. The server's generative AI model (e.g., GPT-4) then generates the optimal answer based on this data. The generated answer includes a list of current events in Kioicho and a description of recommended tourist spots.

[0549] Search result consolidation and submission

[0550] The server integrates the generated answers with map links to restaurants in the Kioicho area and other related map information, including map information obtained from external services such as Google Maps API, to generate the final search result data, which is then sent to the device.

[0551] Displaying search results

[0552] The device displays the received search results to the user. Specifically, it displays information such as a list of events being held in Kioicho, a list of tourist spots, and map links to nearby restaurants. The user can check the displayed information, select events or spots that interest them, and click on map links as needed. This allows the user to obtain information about nearby restaurants.

[0553] Specific examples

[0554] For example, if a user enters the query "Kioicho," the following search results will be displayed:

[0555] List of current events in Kioicho

[0556] "2023 Spring Festival"

[0557] Weekend live music

[0558] "Art Gallery Exhibition"

[0559] Recommended tourist spots

[0560] "Kioicho Garden"

[0561] "History Museum"

[0562] "Shopping mall"

[0563] Map link to nearby restaurants

[0564] "Restaurant A"

[0565] "Cafe B"

[0566] "Bar C"

[0567] Example prompts for generative AI models

[0568] Query: "Kioicho"

[0569] Task for the generative AI: "In response to a vague user query such as 'Kioicho,' generate an answer that includes a list of current events in Kioicho, recommended tourist spots, and map links to nearby restaurants. Relevant information should be obtained from company catalogs and knowledge bases. Map information should be provided using the Google Maps API."

[0570] This invention enables users to quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. In particular, it is expected that the user experience will be significantly improved by providing search results that integrate related map information and event information.

[0571] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0572] Processing Steps

[0573] Step 1:

[0574] Entering a query

[0575] The user enters the query "Kioicho" into the terminal and clicks the search button.

[0576] Specific actions: A user types "Kioicho" into the search bar of a web browser and presses the enter key, or performs a similar operation using a dedicated search app.

[0577] Input: The vague query "Kioicho" entered by the user.

[0578] Output: User input is saved to the terminal.

[0579] Step 2:

[0580] Submitting a query

[0581] The terminal sends the query "Kioicho" entered by the user to the server.

[0582] Specific operation: The device sends an HTTP POST request to the server and includes the query data.

[0583] Input: The device saved query "Kioicho".

[0584] Output: The query is sent to the server.

[0585] Step 3:

[0586] Parsing a query

[0587] The server analyzes the received query and identifies any ambiguities.

[0588] What happens: The server uses a natural language processing (NLP) library (e.g., spaCy or BERT) to tokenize the query and identify grammatical ambiguities.

[0589] Input: The query "Kioicho" received by the server.

[0590] Output: The analysis results in identifying ambiguities.

[0591] Step 4:

[0592] Retrieving related data

[0593] The server retrieves the relevant data from a database maintained by the company.

[0594] Specific operation: The server executes a query against the SQL database to retrieve relevant information, such as "information about current events in Kioicho" or "data about recommended tourist spots."

[0595] Input: A query based on the analysis results.

[0596] Output: Related data obtained (event information, tourist spot information, etc.).

[0597] Step 5:

[0598] Generate answers

[0599] The server's generative AI means (e.g., GPT-4) generates the optimal answer based on the acquired data.

[0600] What it does: The generative AI model takes data as input and creates an appropriate response based on the prompt, for example, listing "All events currently happening in Kioicho."

[0601] Input: The retrieved data and the prompt statement.

[0602] Output: Generated answers (event list, tourist attraction description, etc.).

[0603] Step 6:

[0604] Integration of results

[0605] In addition to the generated answers, the server integrates map links to restaurants around Kioicho and related map information.

[0606] Specific operation: The server calls the Google Maps API to obtain map information for restaurants and tourist spots related to "Kioicho" and includes it in the generated answer.

[0607] Input: Generated answers and map information obtained from the Google Maps API.

[0608] Output: The final consolidated search results.

[0609] Step 7:

[0610] Sending the results

[0611] The server generates the final search result data and transmits it to the terminal.

[0612] Specific operation: The server compiles the generated answer and map information and sends it to the terminal as an HTTP response.

[0613] Input: Consolidated search results data.

[0614] Output: Search result data sent to the device.

[0615] Step 8:

[0616] Displaying the results

[0617] The terminal displays the received search results to the user.

[0618] What it does: The device uses HTML and CSS to render the content it receives as a web page and displays it to the user.

[0619] Input: The received search result data.

[0620] Output: Search results displayed to the user.

[0621] Step 9:

[0622] Verify the information

[0623] The user checks the displayed information, selects events or spots of interest, and clicks on map links if necessary.

[0624] What Happens: A user scrolls through the search results and clicks a link that interests them (e.g., event details or a map). The browser opens the link and displays additional information.

[0625] Input: The search results shown to the user.

[0626] Output: Additional information displayed based on the link the user selects.

[0627] (Application example 1)

[0628] 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."

[0629] In physical stores, it is difficult to provide quick and accurate information in response to vague queries entered by users. This can result in customers not being able to obtain the information they need accurately, which can lead to lower satisfaction. There is also a need to provide a means for customers visiting physical stores to instantly access various information within the store.

[0630] 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.

[0631] In this invention, the server includes means for receiving an ambiguous query entered by a user, a generation AI means for analyzing the ambiguous query and generating an appropriate answer, a means for integrating the generated answer and related information to generate a final search result, a means for transmitting the search result to the user's terminal and displaying it to the user, and a means for allowing the user to obtain search information via hardware located in a physical store. This makes it possible to provide information in response to a user's ambiguous query quickly and accurately even in a physical store, and is expected to improve customer satisfaction.

[0632] An "ambiguous query" is a search request or question entered by a user without a clear, specific intent.

[0633] "Generative AI Means" means a processing means that employs artificial intelligence techniques to generate appropriate responses to received queries.

[0634] A "catalog" refers to a database that lists product information and service details held by a company.

[0635] A "knowledge base" refers to a database that systematically organizes a company's specialized knowledge and past data.

[0636] "Hardware" refers to any physical device or equipment used to process information, including smartphones, smart glasses, and tablet devices.

[0637] "User device" refers to a device that a user directly operates to input and receive information, including smartphones and tablet devices.

[0638] "Means for generating search results" refers to the processing means that aggregates and organizes the final search results based on the answers obtained by the AI ​​generation means in response to ambiguous queries, and provides them to the user.

[0639] "Map information" refers to data containing geographical information about a specific area or location, such as the location of a physical store or nearby facilities.

[0640] "Means for obtaining search information" refers to the technological means that enable users to effectively search for and obtain the information they need in a physical store, including tablet devices and information kiosks installed in the store.

[0641] This invention provides a system for providing quick and accurate information in response to ambiguous queries entered by users in a physical store. The system is implemented using a user terminal, a server, and hardware installed in the physical store.

[0642] System Configuration

[0643] User devices, including smartphones and tablets, where users enter ambiguous queries.

[0644] Server: The server mainly performs the following processes:

[0645] 1. Receiving a query: Receive a query sent from a user terminal.

[0646] 2. Query Analysis: Analyze the ambiguity of the received query and generate an appropriate answer.

[0647] 3. Data Acquisition: Obtain relevant data from catalogs and knowledge bases maintained by the company.

[0648] 4. Generative AI method: Based on this data, the optimal answer is generated using generative AI method.

[0649] 5. Integration and transmission of search results: The generated answers and related information are integrated to generate the final search results and transmit them to the user terminal.

[0650] Processing details

[0651] The server uses OpenAI's GPT-3 model as its generative AI method, which allows it to generate highly accurate answers even for ambiguous queries. Specifically, if a user enters the query "What are the recommended sale items?", the following prompt sentence is input to the generative AI model:

[0652] What sale items do you recommend?

[0653] Catalog: [Sales and product information]

[0654] Knowledge Base: [Store recommendations, customer reviews, sale dates, etc.]

[0655] Use on devices

[0656] Hardware installed in physical stores: Information kiosks and tablet devices are installed in stores, and users can use them to search for information. For example, if a user types "What are some recommended restaurants nearby?" into a tablet device, the server retrieves relevant data and generates the optimal answer using generative AI methods. The generated answer includes a list of nearby restaurants and map information.

[0657] Hardware and Software Use

[0658] The server is implemented using the Python programming language and operates the generative AI model through the OpenAI API. Communication between the user device and the server uses the HTTPS protocol. The hardware placed in the physical store includes tablet devices and information kiosks, which communicate with the server to obtain and display information in real time.

[0659] This system is expected to improve customer satisfaction by allowing users to quickly and accurately obtain the information they need in physical stores.

[0660] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0661] Step 1:

[0662] Receiving a query

[0663] A user enters a vague query on a smartphone or tablet in a physical store, for example, "What are some recommended sale items?"

[0664] Input: A vague query entered by the user

[0665] Output: Query data for the device to send to the server

[0666] Step 2:

[0667] Submitting a query

[0668] The device sends the entered fuzzy query to the server over the HTTPS protocol.

[0669] Input: An ambiguous query typed into the terminal

[0670] Output: Query data received by the server

[0671] Step 3:

[0672] Parsing a query

[0673] The server analyzes the ambiguous query and identifies the ambiguity, for example, what is a "sale item."

[0674] Input: The ambiguous query received by the server

[0675] Output: Parsed query information (e.g., keyword "sale items")

[0676] Step 4:

[0677] Data Acquisition

[0678] The server retrieves the necessary data from the company's catalog and knowledge base.

[0679] Input: Parsed query information

[0680] Output: Catalog and Knowledge Base data

[0681] Step 5:

[0682] Generate prompt statement

[0683] The server generates prompt sentences to input into the generative AI model based on the acquired data.

[0684] Input: Catalog data and knowledge base data

[0685] Output: A prompt to be input to the generative AI model

[0686] Step 6:

[0687] Answer generation using generative AI methods

[0688] The server uses generative AI methods (e.g., OpenAI GPT-3 model) to generate the best answer based on the prompt.

[0689] Input: prompt statement

[0690] Output: Generated answer (e.g. "This week's featured sale items are...")

[0691] Step 7:

[0692] Search result integration

[0693] The server integrates the generated answers with related information (e.g., maps and links) to generate the final search results.

[0694] Input: Generated answers and related information

[0695] Output: Consolidated search results data

[0696] Step 8:

[0697] Submit search results

[0698] The server sends the final search results to the user's terminal.

[0699] Input: Integrated search result data

[0700] Output: Search result data received by the device

[0701] Step 9:

[0702] Displaying search results

[0703] The device then displays the search results to the user, such as a list of sale items or a map link.

[0704] Input: Search result data received by the device

[0705] Output: Search results displayed so that users can see them visually

[0706] These specific processing steps enable users to quickly and accurately obtain the information they need in a physical store.

[0707] 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.

[0708] This invention is a system that uses a generative AI means and an emotion engine to provide appropriate search results for ambiguous queries entered by users. The system includes a user terminal, a server, an emotion engine, and a communication means for linking these. It also uses a catalog and knowledge base maintained by a company to fine-tune the generative AI means.

[0709] Below, the processing flow of the program is shown, and a specific example of operation based on this flow is explained.

[0710] Receiving and parsing queries

[0711] 1. The user enters the query "Kioicho" into the search box on the device and presses the search button.

[0712] 2. The device sends the query "Kioicho" to the server.

[0713] 3. The server analyzes the received query and identifies its ambiguities.

[0714] Generative AI methods and emotion engines generate appropriate answers

[0715] 1. The server accesses the company's catalog database to retrieve the latest event information about Kioicho and related tourist spots. It also retrieves related information from the knowledge base.

[0716] 2. The server's AI generates optimal answers based on the acquired data. The generated answers include a list of events currently being held in Kioicho and descriptions of recommended tourist spots.

[0717] 3. The emotion engine recognizes and analyzes the user's emotions based on their input and past behavioral data. For example, it can determine emotions such as excitement, joy, or sadness based on the user's typing speed and choice of words.

[0718] 4. The server optimizes the response provided by the generative AI means based on the emotional information recognized by the emotion engine. For example, if the user is feeling stressed, the server may recommend relaxing tourist spots.

[0719] Search result consolidation and submission

[0720] 1. In addition to the generated answer, the server integrates map links to restaurants around Kioicho and related map information.

[0721] 2. The server adjusts the display format of search results according to the user's emotions recognized by the emotion engine. For example, if the user feels like relaxing, the server displays the results in a calming interface.

[0722] 3. The server generates the final search result data and sends it to the user's device. The search result data includes the generated answer, event list, tourist attraction list, and map link.

[0723] Displaying search results

[0724] 1. The device displays the received search results to the user. Specifically, it displays information such as a list of events being held in Kioicho, a list of tourist spots, and a map link to nearby restaurants.

[0725] 2. The user reviews the search results and selects the events or spots they are interested in. If necessary, they can click on the map link to obtain more detailed information.

[0726] Specific examples

[0727] For example, if a user enters the query "Kioicho," the following search results will be displayed:

[0728] List of current events in Kioicho

[0729] "2023 Spring Festival"

[0730] Weekend live music

[0731] "Art Gallery Exhibition"

[0732] Recommended tourist spots

[0733] "Kioicho Garden"

[0734] "History Museum"

[0735] "Shopping mall"

[0736] Map link to nearby restaurants

[0737] "Restaurant A"

[0738] "Cafe B"

[0739] "Bar C"

[0740] Furthermore, if the system detects that the user is feeling stressed, it will prioritize tourist spots that are relaxing (e.g., Kioicho Garden) and display the search results in a calming color scheme, providing a user-friendly UI.

[0741] In this way, users can quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. This system is expected to significantly improve user experience.

[0742] The processing flow will be explained below.

[0743] Step 1:

[0744] The user enters the query "Kioicho" into the search box on the device and presses the search button.

[0745] Step 2:

[0746] The terminal sends the query "Kioicho" entered by the user to the server.

[0747] Step 3:

[0748] The server analyzes the received query, determines that it is an ambiguous query, and determines that it needs to use generative AI methods to generate an appropriate answer.

[0749] Step 4:

[0750] The server accesses the company's catalog database to retrieve the latest event information about Kioicho and related tourist spots, and also retrieves related information from the knowledge base.

[0751] Step 5:

[0752] The server's AI generator generates optimal answers based on the acquired data. These answers include a list of events currently being held in Kioicho and descriptions of recommended tourist spots.

[0753] Step 6:

[0754] The emotion engine recognizes and analyzes the user's emotions at that time based on their input, past behavioral data, facial expressions, tone of voice, etc. For example, it can determine emotions such as excitement, joy, or sadness based on the user's input speed and choice of words.

[0755] Step 7:

[0756] The server optimizes the response provided by the generative AI means based on the emotional information recognized by the emotion engine. For example, if the user is feeling stressed, the server will adjust the response to recommend relaxing tourist spots.

[0757] Step 8:

[0758] In addition to the generated answers, the server integrates map links to restaurants around Kioicho and related map information.

[0759] Step 9:

[0760] The server adjusts the display format of search results according to the user's emotions recognized by the emotion engine. For example, if the user feels like relaxing, the server displays the results in a calming interface.

[0761] Step 10:

[0762] The server generates the final search result data and sends it to the user's device. The search result data includes the generated answer, an event list, a list of tourist attractions, and a map link.

[0763] Step 11:

[0764] The device then displays the received search results to the user, such as a list of events currently being held in Kioicho, a list of tourist spots, and map links to nearby restaurants.

[0765] Step 12:

[0766] Users can check the search results and select events or spots they are interested in. If necessary, they can click on a map link to get more detailed information.

[0767] Example 2

[0768] 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."

[0769] Conventional search systems have difficulty providing appropriate answers to ambiguous queries entered by users, and they are unable to provide appropriate search results that reflect the user's emotions, resulting in a poor user experience.

[0770] 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.

[0771] In this invention, the server includes means for receiving an ambiguous query entered by a user, generation AI means for analyzing the ambiguous query and generating an appropriate answer, means for integrating the generated answer and related information to generate final search results, an emotion engine for recognizing and analyzing emotions from the user's input and past behavioral data, and means for optimizing the answer by the generation AI means based on the emotion information recognized by the emotion engine. This allows the user to quickly obtain appropriate search results even for an ambiguous query, and further provides an optimal answer according to the user's emotions, thereby improving the user experience.

[0772] A "query" is a word or phrase that a user enters to search for information.

[0773] A "generative AI means" is an artificial intelligence technology for analyzing a received query and generating an appropriate response to that query.

[0774] The "emotion engine" is a function that recognizes and analyzes emotions from user input and past behavioral data.

[0775] A "database" is a collection of specific information that is systematically organized, stored, and can be accessed as needed.

[0776] A "knowledge base" is a database that systematically organizes and accumulates specific specialized knowledge and information.

[0777] "Cross-reference" is a function or method that allows related information to be mutually referenced.

[0778] The "integration means" is a function for compiling the generated answers and related information into a single result and providing it to the user.

[0779] "Optimization" is the process of adjusting or improving to obtain the best possible condition or result for a particular purpose or condition.

[0780] "Search results" are a collection of answers and related information provided in response to a user-entered query.

[0781] This invention is a system that uses a generative AI means and an emotion engine to provide appropriate search results for ambiguous queries entered by users. The system includes a user terminal, a server, an emotion engine, and a communication means for linking these. Furthermore, the system uses a database and knowledge base maintained by a company to fine-tune the generative AI means.

[0782] System Overview

[0783] User terminal

[0784] A user terminal is a device through which a user enters a query and receives and displays search results. Specific hardware examples include PCs, smartphones, and tablets. Users enter queries and view results through this terminal.

[0785] server

[0786] The server is responsible for parsing queries, generating answers using generative AI means, recognizing and analyzing emotions using an emotion engine, and integrating and transmitting search results. The server is equipped with a high-performance processor and a large amount of memory, allowing for real-time data processing.

[0787] Generation AI means

[0788] Generative AI is used to analyze queries and generate appropriate answers. Specifically, it uses machine learning algorithms to analyze queries and generate optimal answers based on relevant information. This generative AI is fine-tuned using the company's database and knowledge base.

[0789] Emotion Engine

[0790] The emotion engine is an engine that recognizes and analyzes emotions from user input and past behavioral data. Specifically, it analyzes data such as input speed, word choice, and query frequency to determine whether the user is feeling excitement, joy, sadness, stress, or other emotions.

[0791] Specific examples

[0792] An example of the behavior when a user enters the query "Kioicho" into the search box of a terminal will be described.

[0793] 1. Receiving a query: The user enters the query "Kioicho" in the search box of the device and presses the search button. The device sends this query to the server.

[0794] 2. Query Analysis: The server analyzes the received query and identifies its ambiguity. It determines that "Kioicho" is a place name and is associated with a wide range of data, including information about nearby tourist attractions and events.

[0795] 3. Data Acquisition: The server acquires the latest event information and tourist spot data about Kioicho from the company's database and knowledge base.

[0796] 4. Answer generation: Based on the data acquired by the generation AI means, the optimal answer is generated, including descriptions of tourist spots such as "Kioicho Garden" and "Historical Museum," and event information such as "2023 Spring Festival" and "Weekend Music Live."

[0797] 5. Emotion recognition and analysis: The emotion engine recognizes that the user is likely feeling stressed based on their typing speed and choice of words.

[0798] 6. Answer optimization: The server optimizes the answers based on the emotional information recognized by the emotion engine, so that relaxing spots (e.g., "Kioicho Garden") are given priority.

[0799] 7. Integration and transmission of search results: The server integrates the generated answers with map links to restaurants around Kioicho and related map information to generate the final search results and transmit them to the user's device.

[0800] 8. Displaying search results: The device displays the received search results to the user. Specifically, it displays a list of events being held in Kioicho, Kioicho Garden, and a map link to nearby restaurants. The user can check this information and click to view more detailed information as needed.

[0801] In this way, users can quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. Utilizing an emotion engine is expected to significantly improve the quality of the user experience.

[0802] Prompt Sentence Examples

[0803] As an example of a user entering a query, consider the following prompt:

[0804] "Please tell me about historical tourist spots in Kioicho."

[0805] "What events are happening in Kioicho?"

[0806] "Please tell me some recommended restaurants around Kioicho."

[0807] Searches are performed based on these prompts, and generative AI methods and an emotion engine work together to provide optimal search results.

[0808] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0809] Step 1:

[0810] Receiving a query

[0811] The user enters the query "Kioicho" into the search box on the device and presses the search button.

[0812] Input: User-entered query "Kioicho"

[0813] Output: Data sent from the device to the server for the query "Kioicho"

[0814] Specific operation: When the search button on the terminal interface is clicked, query data is constructed and sent to the server.

[0815] Step 2:

[0816] Parsing a query

[0817] The server analyzes the received query and identifies any ambiguities.

[0818] Input: Query "Kioicho" sent from the terminal

[0819] Output: Information about the ambiguity of the query (e.g., it is a place name and may be related to tourist attractions or event information)

[0820] What happens: The server uses natural language processing algorithms to analyze the meaning of the query and identify any ambiguities.

[0821] Step 3:

[0822] Data Acquisition

[0823] The server retrieves the latest event information and tourist spot data about Kioicho from the company's database and knowledge base.

[0824] Input: Query ambiguity information and related database lookup request

[0825] Output: Related event information and tourist spot data (e.g., "2023 Spring Festival," "Kioicho Garden," etc.)

[0826] Specific operation: The server calls the database API, obtains the information corresponding to the query, and stores it in the cache.

[0827] Step 4:

[0828] Generate answers

[0829] The generation AI means generates the optimal answer based on the data acquired.

[0830] Input: Event information and tourist spot data obtained from the database

[0831] Output: A specific answer to be provided to the user (e.g., "Kioicho Garden is a beautiful garden and a great place to relax," "There is live music on weekends," etc.)

[0832] Specific operation: The generative AI means converts the acquired data using a natural language generation algorithm to make it easier for humans to understand.

[0833] Step 5:

[0834] Emotion Recognition and Analysis

[0835] The emotion engine recognizes and analyzes the user's emotions at that time based on their input speed and choice of words.

[0836] Input: User typing speed, words used, and past behavior data

[0837] Output: Information about the user's emotional state (e.g., feeling stressed)

[0838] How it works: The emotion engine uses machine learning algorithms to determine emotions from input data.

[0839] Step 6:

[0840] Optimizing answers

[0841] The server optimizes the answer provided by the generative AI means based on the emotional information recognized by the emotion engine.

[0842] Input: Generated answers and information about the user's emotional state

[0843] Output: Optimized answers (e.g., prioritize relaxation spots for stressed users)

[0844] What happens: The server re-evaluates the answer and adjusts it accordingly.

[0845] Step 7:

[0846] Search result consolidation and submission

[0847] The server then integrates the generated answers with map links to restaurants around Kioicho and other related map information to generate the final search results, which are then sent to the user's device.

[0848] Input: Optimized answer and additional information (e.g. restaurant map link, map information)

[0849] Output: Final search result data

[0850] Specific operation: The server compiles information from various data sources, constructs the final search results, and sends them to the device.

[0851] Step 8:

[0852] Displaying search results

[0853] The terminal displays the received search results to the user.

[0854] Input: Final search result data sent from the server

[0855] Output: Search results displayed on the screen (e.g., "List of events being held in Kioicho," "Kioicho Garden," "Map links to nearby restaurants")

[0856] What happens: The device analyzes the received data and displays it appropriately in the user interface. The user can click on the information of interest to view further details.

[0857] (Application example 2)

[0858] 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."

[0859] Conventional search systems have difficulty providing satisfactory search results when users enter ambiguous queries. Providing optimal search results is particularly challenging when the information desired varies depending on the user's emotions. Food delivery services also face challenges in recommending appropriate restaurants and menus based on ambiguous queries and user emotions.

[0860] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving an ambiguous query entered by a user; a generation AI means for analyzing the ambiguous query and generating an appropriate answer; an emotion engine means for analyzing the user's emotions and optimizing the generated answer and related information; a means for integrating the generated answer and related information to generate final search results; and a means for transmitting the search results to the user's terminal and displaying them to the user. This makes it possible to provide optimal search results that take the user's emotions into consideration, even when the user enters an ambiguous query. Furthermore, in a food delivery service, optimal restaurants and menus can be recommended based on the ambiguous query, thereby improving user satisfaction.

[0861] An "ambiguous query" is an unclear search term that a user enters without providing any specific information.

[0862] "Generative AI methods" refers to artificial intelligence techniques for analyzing ambiguous queries and generating appropriate answers.

[0863] The "emotion engine" is an engine that analyzes emotions from user input and behavioral data and optimizes the generated information.

[0864] "User terminal" refers to the device used by the user to perform a search, including smartphones and personal computers.

[0865] A "corporate catalog" is a database that lists the products and services offered by a company.

[0866] A "knowledge base" is a database that systematically compiles the knowledge and information held by a company.

[0867] "Search Results" refers to answers and related information generated based on a user-entered query.

[0868] "Map information" is data that indicates geographical location information included in the search results.

[0869] "Route information" is data that provides guidance for a user to reach a particular location.

[0870] This invention is a system that provides optimal search results for ambiguous queries entered by users using generative AI means and an emotion engine. This system is embodied as an application called "FoodGenie" that is specialized for food delivery services.

[0871] Program processing

[0872] 1. Receiving a query:

[0873] When a user enters an ambiguous query (e.g., "pizza") on a smartphone app, the device sends this query to a server.

[0874] 2. Emotion analysis:

[0875] The server sends the query to the emotion engine, which analyzes the user's emotional state based on the user's input speed and word choice.

[0876] 3. Answer generation by generative AI:

[0877] The server accesses the company's catalog or knowledge base to retrieve the necessary data, and the generative AI method generates the appropriate answer based on this.

[0878] 4. Emotion-based optimization:

[0879] The generated answers are optimized based on the analysis results from the emotion engine. For example, if the user is feeling stressed, it will recommend a "relaxing family pizza set."

[0880] 5. Search Results Integration:

[0881] The server integrates the generated answers, along with any associated map or route information, to produce the final search results.

[0882] 6. Submitting and Displaying Search Results:

[0883] The server sends the final search results to the user's device and displays them to the user, who then displays the results in the app's interface.

[0884] Hardware and software used

[0885] Device: Smartphone (iOS, Android)

[0886] Emotion Engine API: Emotion-API Platform

[0887] Generate AI model: Custom AI Model API (e.g. OpenAI, GPT-4)

[0888] Server: Cloud server (AWS, Google Cloud)

[0889] Specific examples

[0890] For example, if a user types "pizza" into their smartphone, the emotion engine will detect "stress." The generative AI will prioritize recommendations for restaurants that offer "relaxing family pizza sets."

[0891] Example prompt for a generative AI model:

[0892] plaintext

[0893] A user types "pizza." The user's emotion is "stress." Can you recommend a restaurant and menu that will help them relax and relieve stress?

[0894] This system can provide optimal search results that take emotions into account, even when a user enters an ambiguous query.Furthermore, in food delivery services, it can recommend optimal restaurants and menus based on ambiguous queries, thereby improving user satisfaction.

[0895] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0896] Step 1:

[0897] A user enters a vague query into the "FoodGenie" app on their smartphone and presses the search button. The entered query is a non-specific phrase such as "pizza." Once the input is received, the device sends the query information to the server. The input data includes the query text and the user ID.

[0898] Step 2:

[0899] The server sends a request to the emotion engine API to analyze the received query. The emotion engine takes the query text and user ID as input and analyzes the user's emotional state. For example, it determines the user's emotion (e.g., stress or fatigue) from the query input speed and the words used. The emotion information is returned to the server.

[0900] Step 3:

[0901] The server accesses the company's catalog and knowledge base to gather data related to the query. For example, the server retrieves multiple restaurant and menu data related to "pizza." The results collected from the database include restaurant names, menus, ratings, location information, etc.

[0902] Step 4:

[0903] The server uses a generative AI to analyze the target data and generate the optimal answer. At this time, the server uses the collected restaurant data and the user's emotional information obtained from the emotion engine as input. For example, if it is determined that the user is feeling stressed, the generative AI will suggest a "relaxing family pizza set to relieve stress." The generated answer data includes information such as the restaurant name, recommended menu items, and ratings.

[0904] Step 5:

[0905] The server further integrates the generated answers and refines the final search results. Specifically, it adds map information and route information for the restaurant. It also optimizes the interface based on the analysis results of the emotion engine (for example, displaying calmer colors for users with high stress levels). The final search result data is generated.

[0906] Step 6:

[0907] The server then sends the final search results to the user's device, which then displays them on the app's interface, including the name of the restaurant the user is looking for, map information, and recommended menu items.

[0908] Step 7:

[0909] Users can review the search results displayed in the app and select the restaurant or menu they are interested in. This allows users to easily obtain information and directions to specific restaurants.

[0910] 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.

[0911] 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.

[0912] 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.

[0913] [Third embodiment]

[0914] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0915] 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.

[0916] 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).

[0917] 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.

[0918] 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.

[0919] 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).

[0920] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0921] 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.

[0922] 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.

[0923] 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.

[0924] 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.

[0925] 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."

[0926] This invention is a system that uses a generative AI means to provide appropriate search results for ambiguous queries entered by a user. The system includes a user terminal, a server, and a communication means for linking these. It also uses a catalog and knowledge base maintained by a company to fine-tune the generative AI means.

[0927] Below, the processing flow of the program is shown, and a specific example of operation based on this flow is explained.

[0928] Receiving and parsing queries

[0929] 1. The user enters the search query "Kioicho" into the device and clicks the search button.

[0930] 2. The device sends the query "Kioicho" to the server.

[0931] 3. The server analyzes the received query and identifies its ambiguities.

[0932] Generating appropriate answers using generative AI methods

[0933] 1. The server retrieves relevant data from the company's catalog or knowledge base, such as information about current events in Kioicho or recommended tourist spots.

[0934] 2. The server's AI generator generates optimal answers based on this data. These answers include a list of events currently being held in Kioicho and descriptions of recommended tourist spots.

[0935] Search result consolidation and submission

[0936] 1. In addition to the generated answer, the server integrates map links to restaurants around Kioicho and related map information.

[0937] 2. The server generates the final search result data and sends it to the terminal.

[0938] Displaying search results

[0939] 1. The device displays the received search results to the user. Specifically, it displays information such as a list of events being held in Kioicho, a list of tourist spots, and a map link to nearby restaurants.

[0940] 2. The user checks the displayed information, selects events or spots that interest them, and clicks on map links if necessary, which allows the user to obtain information about nearby restaurants.

[0941] Specific examples

[0942] For example, if a user enters the query "Kioicho," the following search results will be displayed:

[0943] List of current events in Kioicho

[0944] "2023 Spring Festival"

[0945] Weekend live music

[0946] "Art Gallery Exhibition"

[0947] Recommended tourist spots

[0948] "Kioicho Garden"

[0949] "History Museum"

[0950] "Shopping mall"

[0951] Map link to nearby restaurants

[0952] "Restaurant A"

[0953] "Cafe B"

[0954] "Bar C"

[0955] In this way, users can quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. This system is expected to significantly improve user experience.

[0956] The processing flow will be explained below.

[0957] Step 1:

[0958] The user enters the query "Kioicho" into the search box on the device and presses the search button.

[0959] Step 2:

[0960] The terminal sends the query "Kioicho" entered by the user to the server.

[0961] Step 3:

[0962] The server analyzes the received query, determines that it is an ambiguous query, and determines that it needs to use generative AI methods to generate an appropriate answer.

[0963] Step 4:

[0964] The server accesses the company's catalog database to retrieve the latest event information about Kioicho and related tourist spots, and also retrieves related information from the knowledge base.

[0965] Step 5:

[0966] The server's AI generation method generates optimal answers based on the acquired data, such as a list of events currently being held in Kioicho or descriptions of recommended tourist spots.

[0967] Step 6:

[0968] The server will then integrate additional information into the generated answer, specifically including map links to restaurants around Kioicho and related map information in the search results.

[0969] Step 7:

[0970] The server generates the final search result data and sends it to the user's device. The search result data includes the generated answer, an event list, a list of tourist attractions, and a map link.

[0971] Step 8:

[0972] The device analyzes the search results it receives and displays them in an easy-to-understand format for the user, such as a list of events currently being held in Kioicho, a list of tourist spots, and map links to nearby restaurants.

[0973] Step 9:

[0974] Users can check the search results and select events or spots they are interested in. If necessary, they can click on a map link to get more detailed information.

[0975] Example 1

[0976] 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."

[0977] Current search systems have a problem in that they cannot provide sufficiently appropriate results for ambiguous queries entered by users. In particular, when users search without a specific intent, it is difficult to efficiently retrieve relevant information. Furthermore, search results often do not integrate map information or related links, which does not improve the user experience. Therefore, there is a need for a system that can quickly and accurately find the desired information even when users enter ambiguous queries.

[0978] 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.

[0979] In this invention, the server includes means for receiving an ambiguous query entered by a user, means for analyzing the ambiguous query and identifying its ambiguity, means for retrieving related data from a database maintained by the company, means for generating an optimal answer using a generative AI model, means for integrating the generated answer and related map information to generate final search results, and means for transmitting the final search results to the user's terminal and displaying them to the user, thereby enabling the user to quickly and accurately find desired information even for an ambiguous query.

[0980] An "ambiguous query" is a general or vague search term entered by a user without specifying a specific intent or detailed information.

[0981] A "generative AI model" refers to an artificial intelligence that is trained on large datasets and is capable of natural language generation and analysis.

[0982] A "Company-maintained database" is a collection of various sources owned and controlled by the Company that contain information relevant to a particular query.

[0983] A "best answer" refers to the response that provides the most relevant and valuable information to the query entered by the user.

[0984] "Map Information" means information showing locations and routes for a particular geographic area, which may be obtained from external services such as APIs.

[0985] An "HTTP POST request" is one of the communication protocols on the Internet that allows a terminal to send data to a server, and refers to a request that includes the user's query data.

[0986] "Natural language processing (NLP) algorithms" refers to the technologies and methods that enable computers to understand, analyze, and generate human language.

[0987] An "SQL database" is a type of information system for managing and manipulating data using a structured query language.

[0988] "Google Maps API" is an application programming interface for the map information service provided by Google, and is used to obtain specific geographic information.

[0989] "Fine-tuning" refers to additional training to tailor a generative AI model to suit a specific task and data.

[0990] This invention is a system that uses a generative AI means to provide appropriate search results for ambiguous queries entered by a user. The system includes a user terminal, a server, and a communication means for linking these. It also uses a catalog and knowledge base maintained by a company to fine-tune the generative AI means.

[0991] Receiving and parsing queries

[0992] A user enters the query "Kioicho" into a device and clicks the search button. The device sends this query "Kioicho" to the server. The server analyzes the received query and identifies its ambiguity. The analysis uses natural language processing (NLP) algorithms to tokenize the query and identify grammatical ambiguity.

[0993] Generating appropriate answers using generative AI methods

[0994] The server retrieves relevant data from a database maintained by the company. This data includes information on current events in Kioicho and data on recommended tourist spots. An SQL database is used for this retrieval. The server's generative AI model (e.g., GPT-4) then generates the optimal answer based on this data. The generated answer includes a list of current events in Kioicho and a description of recommended tourist spots.

[0995] Search result consolidation and submission

[0996] The server integrates the generated answers with map links to restaurants in the Kioicho area and other related map information, including map information obtained from external services such as Google Maps API, to generate the final search result data, which is then sent to the device.

[0997] Displaying search results

[0998] The device displays the received search results to the user. Specifically, it displays information such as a list of events being held in Kioicho, a list of tourist spots, and map links to nearby restaurants. The user can check the displayed information, select events or spots that interest them, and click on map links as needed. This allows the user to obtain information about nearby restaurants.

[0999] Specific examples

[1000] For example, if a user enters the query "Kioicho," the following search results will be displayed:

[1001] List of current events in Kioicho

[1002] "2023 Spring Festival"

[1003] Weekend live music

[1004] "Art Gallery Exhibition"

[1005] Recommended tourist spots

[1006] "Kioicho Garden"

[1007] "History Museum"

[1008] "Shopping mall"

[1009] Map link to nearby restaurants

[1010] "Restaurant A"

[1011] "Cafe B"

[1012] "Bar C"

[1013] Example prompts for generative AI models

[1014] Query: "Kioicho"

[1015] Task for the generative AI: "In response to a vague user query such as 'Kioicho,' generate an answer that includes a list of current events in Kioicho, recommended tourist spots, and map links to nearby restaurants. Relevant information should be obtained from company catalogs and knowledge bases. Map information should be provided using the Google Maps API."

[1016] This invention enables users to quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. In particular, it is expected that the user experience will be significantly improved by providing search results that integrate related map information and event information.

[1017] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1018] Processing Steps

[1019] Step 1:

[1020] Entering a query

[1021] The user enters the query "Kioicho" into the terminal and clicks the search button.

[1022] Specific actions: A user types "Kioicho" into the search bar of a web browser and presses the enter key, or performs a similar operation using a dedicated search app.

[1023] Input: The vague query "Kioicho" entered by the user.

[1024] Output: User input is saved to the terminal.

[1025] Step 2:

[1026] Submitting a query

[1027] The terminal sends the query "Kioicho" entered by the user to the server.

[1028] Specific operation: The device sends an HTTP POST request to the server and includes the query data.

[1029] Input: The device saved query "Kioicho".

[1030] Output: The query is sent to the server.

[1031] Step 3:

[1032] Parsing a query

[1033] The server analyzes the received query and identifies any ambiguities.

[1034] What happens: The server uses a natural language processing (NLP) library (e.g., spaCy or BERT) to tokenize the query and identify grammatical ambiguities.

[1035] Input: The query "Kioicho" received by the server.

[1036] Output: The analysis results in identifying ambiguities.

[1037] Step 4:

[1038] Retrieving related data

[1039] The server retrieves the relevant data from a database maintained by the company.

[1040] Specific operation: The server executes a query against the SQL database to retrieve relevant information, such as "information about current events in Kioicho" or "data about recommended tourist spots."

[1041] Input: A query based on the analysis results.

[1042] Output: Related data obtained (event information, tourist spot information, etc.).

[1043] Step 5:

[1044] Generate answers

[1045] The server's generative AI means (e.g., GPT-4) generates the optimal answer based on the acquired data.

[1046] What it does: The generative AI model takes data as input and creates an appropriate response based on the prompt, for example, listing "All events currently happening in Kioicho."

[1047] Input: The retrieved data and the prompt statement.

[1048] Output: Generated answers (event list, tourist attraction description, etc.).

[1049] Step 6:

[1050] Integration of results

[1051] In addition to the generated answers, the server integrates map links to restaurants around Kioicho and related map information.

[1052] Specific operation: The server calls the Google Maps API to obtain map information for restaurants and tourist spots related to "Kioicho" and includes it in the generated answer.

[1053] Input: Generated answers and map information obtained from the Google Maps API.

[1054] Output: The final consolidated search results.

[1055] Step 7:

[1056] Sending the results

[1057] The server generates the final search result data and transmits it to the terminal.

[1058] Specific operation: The server compiles the generated answer and map information and sends it to the terminal as an HTTP response.

[1059] Input: Consolidated search results data.

[1060] Output: Search result data sent to the device.

[1061] Step 8:

[1062] Displaying the results

[1063] The terminal displays the received search results to the user.

[1064] What it does: The device uses HTML and CSS to render the content it receives as a web page and displays it to the user.

[1065] Input: The received search result data.

[1066] Output: Search results displayed to the user.

[1067] Step 9:

[1068] Verify the information

[1069] The user checks the displayed information, selects events or spots of interest, and clicks on map links if necessary.

[1070] What Happens: A user scrolls through the search results and clicks a link that interests them (e.g., event details or a map). The browser opens the link and displays additional information.

[1071] Input: The search results shown to the user.

[1072] Output: Additional information displayed based on the link the user selects.

[1073] (Application example 1)

[1074] 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."

[1075] In physical stores, it is difficult to provide quick and accurate information in response to vague queries entered by users. This can result in customers not being able to obtain the information they need accurately, which can lead to lower satisfaction. There is also a need to provide a means for customers visiting physical stores to instantly access various information within the store.

[1076] 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.

[1077] In this invention, the server includes means for receiving an ambiguous query entered by a user, a generation AI means for analyzing the ambiguous query and generating an appropriate answer, a means for integrating the generated answer and related information to generate a final search result, a means for transmitting the search result to the user's terminal and displaying it to the user, and a means for allowing the user to obtain search information via hardware located in a physical store. This makes it possible to provide information in response to a user's ambiguous query quickly and accurately even in a physical store, and is expected to improve customer satisfaction.

[1078] An "ambiguous query" is a search request or question entered by a user without a clear, specific intent.

[1079] "Generative AI Means" means a processing means that employs artificial intelligence techniques to generate appropriate responses to received queries.

[1080] A "catalog" refers to a database that lists product information and service details held by a company.

[1081] A "knowledge base" refers to a database that systematically organizes a company's specialized knowledge and past data.

[1082] "Hardware" refers to any physical device or equipment used to process information, including smartphones, smart glasses, and tablet devices.

[1083] "User device" refers to a device that a user directly operates to input and receive information, including smartphones and tablet devices.

[1084] "Means for generating search results" refers to the processing means that aggregates and organizes the final search results based on the answers obtained by the AI ​​generation means in response to ambiguous queries, and provides them to the user.

[1085] "Map information" refers to data containing geographical information about a specific area or location, such as the location of a physical store or nearby facilities.

[1086] "Means for obtaining search information" refers to the technological means that enable users to effectively search for and obtain the information they need in a physical store, including tablet devices and information kiosks installed in the store.

[1087] This invention provides a system for providing quick and accurate information in response to ambiguous queries entered by users in a physical store. The system is implemented using a user terminal, a server, and hardware installed in the physical store.

[1088] System Configuration

[1089] User devices, including smartphones and tablets, where users enter ambiguous queries.

[1090] Server: The server mainly performs the following processes:

[1091] 1. Receiving a query: Receive a query sent from a user terminal.

[1092] 2. Query Analysis: Analyze the ambiguity of the received query and generate an appropriate answer.

[1093] 3. Data Acquisition: Obtain relevant data from catalogs and knowledge bases maintained by the company.

[1094] 4. Generative AI method: Based on this data, the optimal answer is generated using generative AI method.

[1095] 5. Integration and transmission of search results: The generated answers and related information are integrated to generate the final search results and transmit them to the user terminal.

[1096] Processing details

[1097] The server uses OpenAI's GPT-3 model as its generative AI method, which allows it to generate highly accurate answers even for ambiguous queries. Specifically, if a user enters the query "What are the recommended sale items?", the following prompt sentence is input to the generative AI model:

[1098] What sale items do you recommend?

[1099] Catalog: [Sales and product information]

[1100] Knowledge Base: [Store recommendations, customer reviews, sale dates, etc.]

[1101] Use on devices

[1102] Hardware installed in physical stores: Information kiosks and tablet devices are installed in stores, and users can use them to search for information. For example, if a user types "What are some recommended restaurants nearby?" into a tablet device, the server retrieves relevant data and generates the optimal answer using generative AI methods. The generated answer includes a list of nearby restaurants and map information.

[1103] Hardware and Software Use

[1104] The server is implemented using the Python programming language and operates the generative AI model through the OpenAI API. Communication between the user device and the server uses the HTTPS protocol. The hardware placed in the physical store includes tablet devices and information kiosks, which communicate with the server to obtain and display information in real time.

[1105] This system is expected to improve customer satisfaction by allowing users to quickly and accurately obtain the information they need in physical stores.

[1106] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1107] Step 1:

[1108] Receiving a query

[1109] A user enters a vague query on a smartphone or tablet in a physical store, for example, "What are some recommended sale items?"

[1110] Input: A vague query entered by the user

[1111] Output: Query data for the device to send to the server

[1112] Step 2:

[1113] Submitting a query

[1114] The device sends the entered fuzzy query to the server over the HTTPS protocol.

[1115] Input: An ambiguous query typed into the terminal

[1116] Output: Query data received by the server

[1117] Step 3:

[1118] Parsing a query

[1119] The server analyzes the ambiguous query and identifies the ambiguity, for example, what is a "sale item."

[1120] Input: The ambiguous query received by the server

[1121] Output: Parsed query information (e.g., keyword "sale items")

[1122] Step 4:

[1123] Data Acquisition

[1124] The server retrieves the necessary data from the company's catalog and knowledge base.

[1125] Input: Parsed query information

[1126] Output: Catalog and Knowledge Base data

[1127] Step 5:

[1128] Generate prompt statement

[1129] The server generates prompt sentences to input into the generative AI model based on the acquired data.

[1130] Input: Catalog data and knowledge base data

[1131] Output: A prompt to be input to the generative AI model

[1132] Step 6:

[1133] Answer generation using generative AI methods

[1134] The server uses generative AI methods (e.g., OpenAI GPT-3 model) to generate the best answer based on the prompt.

[1135] Input: prompt statement

[1136] Output: Generated answer (e.g. "This week's featured sale items are...")

[1137] Step 7:

[1138] Search result integration

[1139] The server integrates the generated answers with related information (e.g., maps and links) to generate the final search results.

[1140] Input: Generated answers and related information

[1141] Output: Consolidated search results data

[1142] Step 8:

[1143] Submit search results

[1144] The server sends the final search results to the user's terminal.

[1145] Input: Integrated search result data

[1146] Output: Search result data received by the device

[1147] Step 9:

[1148] Displaying search results

[1149] The device then displays the search results to the user, such as a list of sale items or a map link.

[1150] Input: Search result data received by the device

[1151] Output: Search results displayed so that users can see them visually

[1152] These specific processing steps enable users to quickly and accurately obtain the information they need in a physical store.

[1153] 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.

[1154] This invention is a system that uses a generative AI means and an emotion engine to provide appropriate search results for ambiguous queries entered by users. The system includes a user terminal, a server, an emotion engine, and a communication means for linking these. It also uses a catalog and knowledge base maintained by a company to fine-tune the generative AI means.

[1155] Below, the processing flow of the program is shown, and a specific example of operation based on this flow is explained.

[1156] Receiving and parsing queries

[1157] 1. The user enters the query "Kioicho" into the search box on the device and presses the search button.

[1158] 2. The device sends the query "Kioicho" to the server.

[1159] 3. The server analyzes the received query and identifies its ambiguities.

[1160] Generative AI methods and emotion engines generate appropriate answers

[1161] 1. The server accesses the company's catalog database to retrieve the latest event information about Kioicho and related tourist spots. It also retrieves related information from the knowledge base.

[1162] 2. The server's AI generates optimal answers based on the acquired data. The generated answers include a list of events currently being held in Kioicho and descriptions of recommended tourist spots.

[1163] 3. The emotion engine recognizes and analyzes the user's emotions based on their input and past behavioral data. For example, it can determine emotions such as excitement, joy, or sadness based on the user's typing speed and choice of words.

[1164] 4. The server optimizes the response provided by the generative AI means based on the emotional information recognized by the emotion engine. For example, if the user is feeling stressed, the server may recommend relaxing tourist spots.

[1165] Search result consolidation and submission

[1166] 1. In addition to the generated answer, the server integrates map links to restaurants around Kioicho and related map information.

[1167] 2. The server adjusts the display format of search results according to the user's emotions recognized by the emotion engine. For example, if the user feels like relaxing, the server displays the results in a calming interface.

[1168] 3. The server generates the final search result data and sends it to the user's device. The search result data includes the generated answer, event list, tourist attraction list, and map link.

[1169] Displaying search results

[1170] 1. The device displays the received search results to the user. Specifically, it displays information such as a list of events being held in Kioicho, a list of tourist spots, and a map link to nearby restaurants.

[1171] 2. The user reviews the search results and selects the events or spots they are interested in. If necessary, they can click on the map link to obtain more detailed information.

[1172] Specific examples

[1173] For example, if a user enters the query "Kioicho," the following search results will be displayed:

[1174] List of current events in Kioicho

[1175] "2023 Spring Festival"

[1176] Weekend live music

[1177] "Art Gallery Exhibition"

[1178] Recommended tourist spots

[1179] "Kioicho Garden"

[1180] "History Museum"

[1181] "Shopping mall"

[1182] Map link to nearby restaurants

[1183] "Restaurant A"

[1184] "Cafe B"

[1185] "Bar C"

[1186] Furthermore, if the system detects that the user is feeling stressed, it will prioritize tourist spots that are relaxing (e.g., Kioicho Garden) and display the search results in a calming color scheme, providing a user-friendly UI.

[1187] In this way, users can quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. This system is expected to significantly improve user experience.

[1188] The processing flow will be explained below.

[1189] Step 1:

[1190] The user enters the query "Kioicho" into the search box on the device and presses the search button.

[1191] Step 2:

[1192] The terminal sends the query "Kioicho" entered by the user to the server.

[1193] Step 3:

[1194] The server analyzes the received query, determines that it is an ambiguous query, and determines that it needs to use generative AI methods to generate an appropriate answer.

[1195] Step 4:

[1196] The server accesses the company's catalog database to retrieve the latest event information about Kioicho and related tourist spots, and also retrieves related information from the knowledge base.

[1197] Step 5:

[1198] The server's AI generator generates optimal answers based on the acquired data. These answers include a list of events currently being held in Kioicho and descriptions of recommended tourist spots.

[1199] Step 6:

[1200] The emotion engine recognizes and analyzes the user's emotions at that time based on their input, past behavioral data, facial expressions, tone of voice, etc. For example, it can determine emotions such as excitement, joy, or sadness based on the user's input speed and choice of words.

[1201] Step 7:

[1202] The server optimizes the response provided by the generative AI means based on the emotional information recognized by the emotion engine. For example, if the user is feeling stressed, the server will adjust the response to recommend relaxing tourist spots.

[1203] Step 8:

[1204] In addition to the generated answers, the server integrates map links to restaurants around Kioicho and related map information.

[1205] Step 9:

[1206] The server adjusts the display format of search results according to the user's emotions recognized by the emotion engine. For example, if the user feels like relaxing, the server displays the results in a calming interface.

[1207] Step 10:

[1208] The server generates the final search result data and sends it to the user's device. The search result data includes the generated answer, an event list, a list of tourist attractions, and a map link.

[1209] Step 11:

[1210] The device then displays the received search results to the user, such as a list of events currently being held in Kioicho, a list of tourist spots, and map links to nearby restaurants.

[1211] Step 12:

[1212] Users can check the search results and select events or spots they are interested in. If necessary, they can click on a map link to get more detailed information.

[1213] Example 2

[1214] 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."

[1215] Conventional search systems have difficulty providing appropriate answers to ambiguous queries entered by users, and they are unable to provide appropriate search results that reflect the user's emotions, resulting in a poor user experience.

[1216] 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.

[1217] In this invention, the server includes means for receiving an ambiguous query entered by a user, generation AI means for analyzing the ambiguous query and generating an appropriate answer, means for integrating the generated answer and related information to generate final search results, an emotion engine for recognizing and analyzing emotions from the user's input and past behavioral data, and means for optimizing the answer by the generation AI means based on the emotion information recognized by the emotion engine. This allows the user to quickly obtain appropriate search results even for an ambiguous query, and further provides an optimal answer according to the user's emotions, thereby improving the user experience.

[1218] A "query" is a word or phrase that a user enters to search for information.

[1219] A "generative AI means" is an artificial intelligence technology for analyzing a received query and generating an appropriate response to that query.

[1220] The "emotion engine" is a function that recognizes and analyzes emotions from user input and past behavioral data.

[1221] A "database" is a collection of specific information that is systematically organized, stored, and can be accessed as needed.

[1222] A "knowledge base" is a database that systematically organizes and accumulates specific specialized knowledge and information.

[1223] "Cross-reference" is a function or method that allows related information to be mutually referenced.

[1224] The "integration means" is a function for compiling the generated answers and related information into a single result and providing it to the user.

[1225] "Optimization" is the process of adjusting or improving to obtain the best possible condition or result for a particular purpose or condition.

[1226] "Search results" are a collection of answers and related information provided in response to a user-entered query.

[1227] This invention is a system that uses a generative AI means and an emotion engine to provide appropriate search results for ambiguous queries entered by users. The system includes a user terminal, a server, an emotion engine, and a communication means for linking these. Furthermore, the system uses a database and knowledge base maintained by a company to fine-tune the generative AI means.

[1228] System Overview

[1229] User terminal

[1230] A user terminal is a device through which a user enters a query and receives and displays search results. Specific hardware examples include PCs, smartphones, and tablets. Users enter queries and view results through this terminal.

[1231] server

[1232] The server is responsible for parsing queries, generating answers using generative AI means, recognizing and analyzing emotions using an emotion engine, and integrating and transmitting search results. The server is equipped with a high-performance processor and a large amount of memory, allowing for real-time data processing.

[1233] Generation AI means

[1234] Generative AI is used to analyze queries and generate appropriate answers. Specifically, it uses machine learning algorithms to analyze queries and generate optimal answers based on relevant information. This generative AI is fine-tuned using the company's database and knowledge base.

[1235] Emotion Engine

[1236] The emotion engine is an engine that recognizes and analyzes emotions from user input and past behavioral data. Specifically, it analyzes data such as input speed, word choice, and query frequency to determine whether the user is feeling excitement, joy, sadness, stress, or other emotions.

[1237] Specific examples

[1238] An example of the behavior when a user enters the query "Kioicho" into the search box of a terminal will be described.

[1239] 1. Receiving a query: The user enters the query "Kioicho" in the search box of the device and presses the search button. The device sends this query to the server.

[1240] 2. Query Analysis: The server analyzes the received query and identifies its ambiguity. It determines that "Kioicho" is a place name and is associated with a wide range of data, including information about nearby tourist attractions and events.

[1241] 3. Data Acquisition: The server acquires the latest event information and tourist spot data about Kioicho from the company's database and knowledge base.

[1242] 4. Answer generation: Based on the data acquired by the generation AI means, the optimal answer is generated, including descriptions of tourist spots such as "Kioicho Garden" and "Historical Museum," and event information such as "2023 Spring Festival" and "Weekend Music Live."

[1243] 5. Emotion recognition and analysis: The emotion engine recognizes that the user is likely feeling stressed based on their typing speed and choice of words.

[1244] 6. Answer optimization: The server optimizes the answers based on the emotional information recognized by the emotion engine, so that relaxing spots (e.g., "Kioicho Garden") are given priority.

[1245] 7. Integration and transmission of search results: The server integrates the generated answers with map links to restaurants around Kioicho and related map information to generate the final search results and transmit them to the user's device.

[1246] 8. Displaying search results: The device displays the received search results to the user. Specifically, it displays a list of events being held in Kioicho, Kioicho Garden, and a map link to nearby restaurants. The user can check this information and click to view more detailed information as needed.

[1247] In this way, users can quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. Utilizing an emotion engine is expected to significantly improve the quality of the user experience.

[1248] Prompt Sentence Examples

[1249] As an example of a user entering a query, consider the following prompt:

[1250] "Please tell me about historical tourist spots in Kioicho."

[1251] "What events are happening in Kioicho?"

[1252] "Please tell me some recommended restaurants around Kioicho."

[1253] Searches are performed based on these prompts, and generative AI methods and an emotion engine work together to provide optimal search results.

[1254] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1255] Step 1:

[1256] Receiving a query

[1257] The user enters the query "Kioicho" into the search box on the device and presses the search button.

[1258] Input: User-entered query "Kioicho"

[1259] Output: Data sent from the device to the server for the query "Kioicho"

[1260] Specific operation: When the search button on the terminal interface is clicked, query data is constructed and sent to the server.

[1261] Step 2:

[1262] Parsing a query

[1263] The server analyzes the received query and identifies any ambiguities.

[1264] Input: Query "Kioicho" sent from the terminal

[1265] Output: Information about the ambiguity of the query (e.g., it is a place name and may be related to tourist attractions or event information)

[1266] What happens: The server uses natural language processing algorithms to analyze the meaning of the query and identify any ambiguities.

[1267] Step 3:

[1268] Data Acquisition

[1269] The server retrieves the latest event information and tourist spot data about Kioicho from the company's database and knowledge base.

[1270] Input: Query ambiguity information and related database lookup request

[1271] Output: Related event information and tourist spot data (e.g., "2023 Spring Festival," "Kioicho Garden," etc.)

[1272] Specific operation: The server calls the database API, obtains the information corresponding to the query, and stores it in the cache.

[1273] Step 4:

[1274] Generate answers

[1275] The generation AI means generates the optimal answer based on the data acquired.

[1276] Input: Event information and tourist spot data obtained from the database

[1277] Output: A specific answer to be provided to the user (e.g., "Kioicho Garden is a beautiful garden and a great place to relax," "There is live music on weekends," etc.)

[1278] Specific operation: The generative AI means converts the acquired data using a natural language generation algorithm to make it easier for humans to understand.

[1279] Step 5:

[1280] Emotion Recognition and Analysis

[1281] The emotion engine recognizes and analyzes the user's emotions at that time based on their input speed and choice of words.

[1282] Input: User typing speed, words used, and past behavior data

[1283] Output: Information about the user's emotional state (e.g., feeling stressed)

[1284] How it works: The emotion engine uses machine learning algorithms to determine emotions from input data.

[1285] Step 6:

[1286] Optimizing answers

[1287] The server optimizes the answer provided by the generative AI means based on the emotional information recognized by the emotion engine.

[1288] Input: Generated answers and information about the user's emotional state

[1289] Output: Optimized answers (e.g., prioritize relaxation spots for stressed users)

[1290] What happens: The server re-evaluates the answer and adjusts it accordingly.

[1291] Step 7:

[1292] Search result consolidation and submission

[1293] The server then integrates the generated answers with map links to restaurants around Kioicho and other related map information to generate the final search results, which are then sent to the user's device.

[1294] Input: Optimized answer and additional information (e.g. restaurant map link, map information)

[1295] Output: Final search result data

[1296] Specific operation: The server compiles information from various data sources, constructs the final search results, and sends them to the device.

[1297] Step 8:

[1298] Displaying search results

[1299] The terminal displays the received search results to the user.

[1300] Input: Final search result data sent from the server

[1301] Output: Search results displayed on the screen (e.g., "List of events being held in Kioicho," "Kioicho Garden," "Map links to nearby restaurants")

[1302] What happens: The device analyzes the received data and displays it appropriately in the user interface. The user can click on the information of interest to view further details.

[1303] (Application example 2)

[1304] 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."

[1305] Conventional search systems have difficulty providing satisfactory search results when users enter ambiguous queries. Providing optimal search results is particularly challenging when the information desired varies depending on the user's emotions. Food delivery services also face challenges in recommending appropriate restaurants and menus based on ambiguous queries and user emotions.

[1306] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving an ambiguous query entered by a user; a generation AI means for analyzing the ambiguous query and generating an appropriate answer; an emotion engine means for analyzing the user's emotions and optimizing the generated answer and related information; a means for integrating the generated answer and related information to generate final search results; and a means for transmitting the search results to the user's terminal and displaying them to the user. This makes it possible to provide optimal search results that take the user's emotions into consideration, even when the user enters an ambiguous query. Furthermore, in a food delivery service, optimal restaurants and menus can be recommended based on the ambiguous query, thereby improving user satisfaction.

[1307] An "ambiguous query" is an unclear search term that a user enters without providing any specific information.

[1308] "Generative AI methods" refers to artificial intelligence techniques for analyzing ambiguous queries and generating appropriate answers.

[1309] The "emotion engine" is an engine that analyzes emotions from user input and behavioral data and optimizes the generated information.

[1310] "User terminal" refers to the device used by the user to perform a search, including smartphones and personal computers.

[1311] A "corporate catalog" is a database that lists the products and services offered by a company.

[1312] A "knowledge base" is a database that systematically compiles the knowledge and information held by a company.

[1313] "Search Results" refers to answers and related information generated based on a user-entered query.

[1314] "Map information" is data that indicates geographical location information included in the search results.

[1315] "Route information" is data that provides guidance for a user to reach a particular location.

[1316] This invention is a system that provides optimal search results for ambiguous queries entered by users using generative AI means and an emotion engine. This system is embodied as an application called "FoodGenie" that is specialized for food delivery services.

[1317] Program processing

[1318] 1. Receiving a query:

[1319] When a user enters an ambiguous query (e.g., "pizza") on a smartphone app, the device sends this query to a server.

[1320] 2. Emotion analysis:

[1321] The server sends the query to the emotion engine, which analyzes the user's emotional state based on the user's input speed and word choice.

[1322] 3. Answer generation by generative AI:

[1323] The server accesses the company's catalog or knowledge base to retrieve the necessary data, and the generative AI method generates the appropriate answer based on this.

[1324] 4. Emotion-based optimization:

[1325] The generated answers are optimized based on the analysis results from the emotion engine. For example, if the user is feeling stressed, it will recommend a "relaxing family pizza set."

[1326] 5. Search Results Integration:

[1327] The server integrates the generated answers, along with any associated map or route information, to produce the final search results.

[1328] 6. Submitting and Displaying Search Results:

[1329] The server sends the final search results to the user's device and displays them to the user, who then displays the results in the app's interface.

[1330] Hardware and software used

[1331] Device: Smartphone (iOS, Android)

[1332] Emotion Engine API: Emotion-API Platform

[1333] Generate AI model: Custom AI Model API (e.g. OpenAI, GPT-4)

[1334] Server: Cloud server (AWS, Google Cloud)

[1335] Specific examples

[1336] For example, if a user types "pizza" into their smartphone, the emotion engine will detect "stress." The generative AI will prioritize recommendations for restaurants that offer "relaxing family pizza sets."

[1337] Example prompt for a generative AI model:

[1338] plaintext

[1339] A user types "pizza." The user's emotion is "stress." Can you recommend a restaurant and menu that will help them relax and relieve stress?

[1340] This system can provide optimal search results that take emotions into account, even when a user enters an ambiguous query.Furthermore, in food delivery services, it can recommend optimal restaurants and menus based on ambiguous queries, thereby improving user satisfaction.

[1341] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1342] Step 1:

[1343] A user enters a vague query into the "FoodGenie" app on their smartphone and presses the search button. The entered query is a non-specific phrase such as "pizza." Once the input is received, the device sends the query information to the server. The input data includes the query text and the user ID.

[1344] Step 2:

[1345] The server sends a request to the emotion engine API to analyze the received query. The emotion engine takes the query text and user ID as input and analyzes the user's emotional state. For example, it determines the user's emotion (e.g., stress or fatigue) from the query input speed and the words used. The emotion information is returned to the server.

[1346] Step 3:

[1347] The server accesses the company's catalog and knowledge base to gather data related to the query. For example, the server retrieves multiple restaurant and menu data related to "pizza." The results collected from the database include restaurant names, menus, ratings, location information, etc.

[1348] Step 4:

[1349] The server uses a generative AI to analyze the target data and generate the optimal answer. At this time, the server uses the collected restaurant data and the user's emotional information obtained from the emotion engine as input. For example, if it is determined that the user is feeling stressed, the generative AI will suggest a "relaxing family pizza set to relieve stress." The generated answer data includes information such as the restaurant name, recommended menu items, and ratings.

[1350] Step 5:

[1351] The server further integrates the generated answers and refines the final search results. Specifically, it adds map information and route information for the restaurant. It also optimizes the interface based on the analysis results of the emotion engine (for example, displaying calmer colors for users with high stress levels). The final search result data is generated.

[1352] Step 6:

[1353] The server then sends the final search results to the user's device, which then displays them on the app's interface, including the name of the restaurant the user is looking for, map information, and recommended menu items.

[1354] Step 7:

[1355] Users can review the search results displayed in the app and select the restaurant or menu they are interested in. This allows users to easily obtain information and directions to specific restaurants.

[1356] 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.

[1357] 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.

[1358] 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.

[1359] [Fourth embodiment]

[1360] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1361] 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.

[1362] 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).

[1363] 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.

[1364] 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.

[1365] 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).

[1366] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1367] 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.

[1368] 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.

[1369] 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.

[1370] 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.

[1371] 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.

[1372] 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."

[1373] This invention is a system that uses a generative AI means to provide appropriate search results for ambiguous queries entered by a user. The system includes a user terminal, a server, and a communication means for linking these. It also uses a catalog and knowledge base maintained by a company to fine-tune the generative AI means.

[1374] Below, the processing flow of the program is shown, and a specific example of operation based on this flow is explained.

[1375] Receiving and parsing queries

[1376] 1. The user enters the search query "Kioicho" into the device and clicks the search button.

[1377] 2. The device sends the query "Kioicho" to the server.

[1378] 3. The server analyzes the received query and identifies its ambiguities.

[1379] Generating appropriate answers using generative AI methods

[1380] 1. The server retrieves relevant data from the company's catalog or knowledge base, such as information about current events in Kioicho or recommended tourist spots.

[1381] 2. The server's AI generator generates optimal answers based on this data. These answers include a list of events currently being held in Kioicho and descriptions of recommended tourist spots.

[1382] Search result consolidation and submission

[1383] 1. In addition to the generated answer, the server integrates map links to restaurants around Kioicho and related map information.

[1384] 2. The server generates the final search result data and sends it to the terminal.

[1385] Displaying search results

[1386] 1. The device displays the received search results to the user. Specifically, it displays information such as a list of events being held in Kioicho, a list of tourist spots, and a map link to nearby restaurants.

[1387] 2. The user checks the displayed information, selects events or spots that interest them, and clicks on map links if necessary, which allows the user to obtain information about nearby restaurants.

[1388] Specific examples

[1389] For example, if a user enters the query "Kioicho," the following search results will be displayed:

[1390] List of current events in Kioicho

[1391] "2023 Spring Festival"

[1392] Weekend live music

[1393] "Art Gallery Exhibition"

[1394] Recommended tourist spots

[1395] "Kioicho Garden"

[1396] "History Museum"

[1397] "Shopping mall"

[1398] Map link to nearby restaurants

[1399] "Restaurant A"

[1400] "Cafe B"

[1401] "Bar C"

[1402] In this way, users can quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. This system is expected to significantly improve user experience.

[1403] The processing flow will be explained below.

[1404] Step 1:

[1405] The user enters the query "Kioicho" into the search box on the device and presses the search button.

[1406] Step 2:

[1407] The terminal sends the query "Kioicho" entered by the user to the server.

[1408] Step 3:

[1409] The server analyzes the received query, determines that it is an ambiguous query, and determines that it needs to use generative AI methods to generate an appropriate answer.

[1410] Step 4:

[1411] The server accesses the company's catalog database to retrieve the latest event information about Kioicho and related tourist spots, and also retrieves related information from the knowledge base.

[1412] Step 5:

[1413] The server's AI generation method generates optimal answers based on the acquired data, such as a list of events currently being held in Kioicho or descriptions of recommended tourist spots.

[1414] Step 6:

[1415] The server will then integrate additional information into the generated answer, specifically including map links to restaurants around Kioicho and related map information in the search results.

[1416] Step 7:

[1417] The server generates the final search result data and sends it to the user's device. The search result data includes the generated answer, an event list, a list of tourist attractions, and a map link.

[1418] Step 8:

[1419] The device analyzes the search results it receives and displays them in an easy-to-understand format for the user, such as a list of events currently being held in Kioicho, a list of tourist spots, and map links to nearby restaurants.

[1420] Step 9:

[1421] Users can check the search results and select events or spots they are interested in. If necessary, they can click on a map link to get more detailed information.

[1422] Example 1

[1423] 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."

[1424] Current search systems have a problem in that they cannot provide sufficiently appropriate results for ambiguous queries entered by users. In particular, when users search without a specific intent, it is difficult to efficiently retrieve relevant information. Furthermore, search results often do not integrate map information or related links, which does not improve the user experience. Therefore, there is a need for a system that can quickly and accurately find the desired information even when users enter ambiguous queries.

[1425] 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.

[1426] In this invention, the server includes means for receiving an ambiguous query entered by a user, means for analyzing the ambiguous query and identifying its ambiguity, means for retrieving related data from a database maintained by the company, means for generating an optimal answer using a generative AI model, means for integrating the generated answer and related map information to generate final search results, and means for transmitting the final search results to the user's terminal and displaying them to the user, thereby enabling the user to quickly and accurately find desired information even for an ambiguous query.

[1427] An "ambiguous query" is a general or vague search term entered by a user without specifying a specific intent or detailed information.

[1428] A "generative AI model" refers to an artificial intelligence that is trained on large datasets and is capable of natural language generation and analysis.

[1429] A "Company-maintained database" is a collection of various sources owned and controlled by the Company that contain information relevant to a particular query.

[1430] A "best answer" refers to the response that provides the most relevant and valuable information to the query entered by the user.

[1431] "Map Information" means information showing locations and routes for a particular geographic area, which may be obtained from external services such as APIs.

[1432] An "HTTP POST request" is one of the communication protocols on the Internet that allows a terminal to send data to a server, and refers to a request that includes the user's query data.

[1433] "Natural language processing (NLP) algorithms" refers to the technologies and methods that enable computers to understand, analyze, and generate human language.

[1434] An "SQL database" is a type of information system for managing and manipulating data using a structured query language.

[1435] "Google Maps API" is an application programming interface for the map information service provided by Google, and is used to obtain specific geographic information.

[1436] "Fine-tuning" refers to additional training to tailor a generative AI model to suit a specific task and data.

[1437] This invention is a system that uses a generative AI means to provide appropriate search results for ambiguous queries entered by a user. The system includes a user terminal, a server, and a communication means for linking these. It also uses a catalog and knowledge base maintained by a company to fine-tune the generative AI means.

[1438] Receiving and parsing queries

[1439] A user enters the query "Kioicho" into a device and clicks the search button. The device sends this query "Kioicho" to the server. The server analyzes the received query and identifies its ambiguity. The analysis uses natural language processing (NLP) algorithms to tokenize the query and identify grammatical ambiguity.

[1440] Generating appropriate answers using generative AI methods

[1441] The server retrieves relevant data from a database maintained by the company. This data includes information on current events in Kioicho and data on recommended tourist spots. An SQL database is used for this retrieval. The server's generative AI model (e.g., GPT-4) then generates the optimal answer based on this data. The generated answer includes a list of current events in Kioicho and a description of recommended tourist spots.

[1442] Search result consolidation and submission

[1443] The server integrates the generated answers with map links to restaurants in the Kioicho area and other related map information, including map information obtained from external services such as Google Maps API, to generate the final search result data, which is then sent to the device.

[1444] Displaying search results

[1445] The device displays the received search results to the user. Specifically, it displays information such as a list of events being held in Kioicho, a list of tourist spots, and map links to nearby restaurants. The user can check the displayed information, select events or spots that interest them, and click on map links as needed. This allows the user to obtain information about nearby restaurants.

[1446] Specific examples

[1447] For example, if a user enters the query "Kioicho," the following search results will be displayed:

[1448] List of current events in Kioicho

[1449] "2023 Spring Festival"

[1450] Weekend live music

[1451] "Art Gallery Exhibition"

[1452] Recommended tourist spots

[1453] "Kioicho Garden"

[1454] "History Museum"

[1455] "Shopping mall"

[1456] Map link to nearby restaurants

[1457] "Restaurant A"

[1458] "Cafe B"

[1459] "Bar C"

[1460] Example prompts for generative AI models

[1461] Query: "Kioicho"

[1462] Task for the generative AI: "In response to a vague user query such as 'Kioicho,' generate an answer that includes a list of current events in Kioicho, recommended tourist spots, and map links to nearby restaurants. Relevant information should be obtained from company catalogs and knowledge bases. Map information should be provided using the Google Maps API."

[1463] This invention enables users to quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. In particular, it is expected that the user experience will be significantly improved by providing search results that integrate related map information and event information.

[1464] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1465] Processing Steps

[1466] Step 1:

[1467] Entering a query

[1468] The user enters the query "Kioicho" into the terminal and clicks the search button.

[1469] Specific actions: A user types "Kioicho" into the search bar of a web browser and presses the enter key, or performs a similar operation using a dedicated search app.

[1470] Input: The vague query "Kioicho" entered by the user.

[1471] Output: User input is saved to the terminal.

[1472] Step 2:

[1473] Submitting a query

[1474] The terminal sends the query "Kioicho" entered by the user to the server.

[1475] Specific operation: The device sends an HTTP POST request to the server and includes the query data.

[1476] Input: The device saved query "Kioicho".

[1477] Output: The query is sent to the server.

[1478] Step 3:

[1479] Parsing a query

[1480] The server analyzes the received query and identifies any ambiguities.

[1481] What happens: The server uses a natural language processing (NLP) library (e.g., spaCy or BERT) to tokenize the query and identify grammatical ambiguities.

[1482] Input: The query "Kioicho" received by the server.

[1483] Output: The analysis results in identifying ambiguities.

[1484] Step 4:

[1485] Retrieving related data

[1486] The server retrieves the relevant data from a database maintained by the company.

[1487] Specific operation: The server executes a query against the SQL database to retrieve relevant information, such as "information about current events in Kioicho" or "data about recommended tourist spots."

[1488] Input: A query based on the analysis results.

[1489] Output: Related data obtained (event information, tourist spot information, etc.).

[1490] Step 5:

[1491] Generate answers

[1492] The server's generative AI means (e.g., GPT-4) generates the optimal answer based on the acquired data.

[1493] What it does: The generative AI model takes data as input and creates an appropriate response based on the prompt, for example, listing "All events currently happening in Kioicho."

[1494] Input: The retrieved data and the prompt statement.

[1495] Output: Generated answers (event list, tourist attraction description, etc.).

[1496] Step 6:

[1497] Integration of results

[1498] In addition to the generated answers, the server integrates map links to restaurants around Kioicho and related map information.

[1499] Specific operation: The server calls the Google Maps API to obtain map information for restaurants and tourist spots related to "Kioicho" and includes it in the generated answer.

[1500] Input: Generated answers and map information obtained from the Google Maps API.

[1501] Output: The final consolidated search results.

[1502] Step 7:

[1503] Sending the results

[1504] The server generates the final search result data and transmits it to the terminal.

[1505] Specific operation: The server compiles the generated answer and map information and sends it to the terminal as an HTTP response.

[1506] Input: Consolidated search results data.

[1507] Output: Search result data sent to the device.

[1508] Step 8:

[1509] Displaying the results

[1510] The terminal displays the received search results to the user.

[1511] What it does: The device uses HTML and CSS to render the content it receives as a web page and displays it to the user.

[1512] Input: The received search result data.

[1513] Output: Search results displayed to the user.

[1514] Step 9:

[1515] Verify the information

[1516] The user checks the displayed information, selects events or spots of interest, and clicks on map links if necessary.

[1517] What Happens: A user scrolls through the search results and clicks a link that interests them (e.g., event details or a map). The browser opens the link and displays additional information.

[1518] Input: The search results shown to the user.

[1519] Output: Additional information displayed based on the link the user selects.

[1520] (Application example 1)

[1521] 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."

[1522] In physical stores, it is difficult to provide quick and accurate information in response to vague queries entered by users. This can result in customers not being able to obtain the information they need accurately, which can lead to lower satisfaction. There is also a need to provide a means for customers visiting physical stores to instantly access various information within the store.

[1523] 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.

[1524] In this invention, the server includes means for receiving an ambiguous query entered by a user, a generation AI means for analyzing the ambiguous query and generating an appropriate answer, a means for integrating the generated answer and related information to generate a final search result, a means for transmitting the search result to the user's terminal and displaying it to the user, and a means for allowing the user to obtain search information via hardware located in a physical store. This makes it possible to provide information in response to a user's ambiguous query quickly and accurately even in a physical store, and is expected to improve customer satisfaction.

[1525] An "ambiguous query" is a search request or question entered by a user without a clear, specific intent.

[1526] "Generative AI Means" means a processing means that employs artificial intelligence techniques to generate appropriate responses to received queries.

[1527] A "catalog" refers to a database that lists product information and service details held by a company.

[1528] A "knowledge base" refers to a database that systematically organizes a company's specialized knowledge and past data.

[1529] "Hardware" refers to any physical device or equipment used to process information, including smartphones, smart glasses, and tablet devices.

[1530] "User device" refers to a device that a user directly operates to input and receive information, including smartphones and tablet devices.

[1531] "Means for generating search results" refers to the processing means that aggregates and organizes the final search results based on the answers obtained by the AI ​​generation means in response to ambiguous queries, and provides them to the user.

[1532] "Map information" refers to data containing geographical information about a specific area or location, such as the location of a physical store or nearby facilities.

[1533] "Means for obtaining search information" refers to the technological means that enable users to effectively search for and obtain the information they need in a physical store, including tablet devices and information kiosks installed in the store.

[1534] This invention provides a system for providing quick and accurate information in response to ambiguous queries entered by users in a physical store. The system is implemented using a user terminal, a server, and hardware installed in the physical store.

[1535] System Configuration

[1536] User devices, including smartphones and tablets, where users enter ambiguous queries.

[1537] Server: The server mainly performs the following processes:

[1538] 1. Receiving a query: Receive a query sent from a user terminal.

[1539] 2. Query Analysis: Analyze the ambiguity of the received query and generate an appropriate answer.

[1540] 3. Data Acquisition: Obtain relevant data from catalogs and knowledge bases maintained by the company.

[1541] 4. Generative AI method: Based on this data, the optimal answer is generated using generative AI method.

[1542] 5. Integration and transmission of search results: The generated answers and related information are integrated to generate the final search results and transmit them to the user terminal.

[1543] Processing details

[1544] The server uses OpenAI's GPT-3 model as its generative AI method, which allows it to generate highly accurate answers even for ambiguous queries. Specifically, if a user enters the query "What are the recommended sale items?", the following prompt sentence is input to the generative AI model:

[1545] What sale items do you recommend?

[1546] Catalog: [Sales and product information]

[1547] Knowledge Base: [Store recommendations, customer reviews, sale dates, etc.]

[1548] Use on devices

[1549] Hardware installed in physical stores: Information kiosks and tablet devices are installed in stores, and users can use them to search for information. For example, if a user types "What are some recommended restaurants nearby?" into a tablet device, the server retrieves relevant data and generates the optimal answer using generative AI methods. The generated answer includes a list of nearby restaurants and map information.

[1550] Hardware and Software Use

[1551] The server is implemented using the Python programming language and operates the generative AI model through the OpenAI API. Communication between the user device and the server uses the HTTPS protocol. The hardware placed in the physical store includes tablet devices and information kiosks, which communicate with the server to obtain and display information in real time.

[1552] This system is expected to improve customer satisfaction by allowing users to quickly and accurately obtain the information they need in physical stores.

[1553] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1554] Step 1:

[1555] Receiving a query

[1556] A user enters a vague query on a smartphone or tablet in a physical store, for example, "What are some recommended sale items?"

[1557] Input: A vague query entered by the user

[1558] Output: Query data for the device to send to the server

[1559] Step 2:

[1560] Submitting a query

[1561] The device sends the entered fuzzy query to the server over the HTTPS protocol.

[1562] Input: An ambiguous query typed into the terminal

[1563] Output: Query data received by the server

[1564] Step 3:

[1565] Parsing a query

[1566] The server analyzes the ambiguous query and identifies the ambiguity, for example, what is a "sale item."

[1567] Input: The ambiguous query received by the server

[1568] Output: Parsed query information (e.g., keyword "sale items")

[1569] Step 4:

[1570] Data Acquisition

[1571] The server retrieves the necessary data from the company's catalog and knowledge base.

[1572] Input: Parsed query information

[1573] Output: Catalog and Knowledge Base data

[1574] Step 5:

[1575] Generate prompt statement

[1576] The server generates prompt sentences to input into the generative AI model based on the acquired data.

[1577] Input: Catalog data and knowledge base data

[1578] Output: A prompt to be input to the generative AI model

[1579] Step 6:

[1580] Answer generation using generative AI methods

[1581] The server uses generative AI methods (e.g., OpenAI GPT-3 model) to generate the best answer based on the prompt.

[1582] Input: prompt statement

[1583] Output: Generated answer (e.g. "This week's featured sale items are...")

[1584] Step 7:

[1585] Search result integration

[1586] The server integrates the generated answers with related information (e.g., maps and links) to generate the final search results.

[1587] Input: Generated answers and related information

[1588] Output: Consolidated search results data

[1589] Step 8:

[1590] Submit search results

[1591] The server sends the final search results to the user's terminal.

[1592] Input: Integrated search result data

[1593] Output: Search result data received by the device

[1594] Step 9:

[1595] Displaying search results

[1596] The device then displays the search results to the user, such as a list of sale items or a map link.

[1597] Input: Search result data received by the device

[1598] Output: Search results displayed so that users can see them visually

[1599] These specific processing steps enable users to quickly and accurately obtain the information they need in a physical store.

[1600] 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.

[1601] This invention is a system that uses a generative AI means and an emotion engine to provide appropriate search results for ambiguous queries entered by users. The system includes a user terminal, a server, an emotion engine, and a communication means for linking these. It also uses a catalog and knowledge base maintained by a company to fine-tune the generative AI means.

[1602] Below, the processing flow of the program is shown, and a specific example of operation based on this flow is explained.

[1603] Receiving and parsing queries

[1604] 1. The user enters the query "Kioicho" into the search box on the device and presses the search button.

[1605] 2. The device sends the query "Kioicho" to the server.

[1606] 3. The server analyzes the received query and identifies its ambiguities.

[1607] Generative AI methods and emotion engines generate appropriate answers

[1608] 1. The server accesses the company's catalog database to retrieve the latest event information about Kioicho and related tourist spots. It also retrieves related information from the knowledge base.

[1609] 2. The server's AI generates optimal answers based on the acquired data. The generated answers include a list of events currently being held in Kioicho and descriptions of recommended tourist spots.

[1610] 3. The emotion engine recognizes and analyzes the user's emotions based on their input and past behavioral data. For example, it can determine emotions such as excitement, joy, or sadness based on the user's typing speed and choice of words.

[1611] 4. The server optimizes the response provided by the generative AI means based on the emotional information recognized by the emotion engine. For example, if the user is feeling stressed, the server may recommend relaxing tourist spots.

[1612] Search result consolidation and submission

[1613] 1. In addition to the generated answer, the server integrates map links to restaurants around Kioicho and related map information.

[1614] 2. The server adjusts the display format of search results according to the user's emotions recognized by the emotion engine. For example, if the user feels like relaxing, the server displays the results in a calming interface.

[1615] 3. The server generates the final search result data and sends it to the user's device. The search result data includes the generated answer, event list, tourist attraction list, and map link.

[1616] Displaying search results

[1617] 1. The device displays the received search results to the user. Specifically, it displays information such as a list of events being held in Kioicho, a list of tourist spots, and a map link to nearby restaurants.

[1618] 2. The user reviews the search results and selects the events or spots they are interested in. If necessary, they can click on the map link to obtain more detailed information.

[1619] Specific examples

[1620] For example, if a user enters the query "Kioicho," the following search results will be displayed:

[1621] List of current events in Kioicho

[1622] "2023 Spring Festival"

[1623] Weekend live music

[1624] "Art Gallery Exhibition"

[1625] Recommended tourist spots

[1626] "Kioicho Garden"

[1627] "History Museum"

[1628] "Shopping mall"

[1629] Map link to nearby restaurants

[1630] "Restaurant A"

[1631] "Cafe B"

[1632] "Bar C"

[1633] Furthermore, if the system detects that the user is feeling stressed, it will prioritize tourist spots that are relaxing (e.g., Kioicho Garden) and display the search results in a calming color scheme, providing a user-friendly UI.

[1634] In this way, users can quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. This system is expected to significantly improve user experience.

[1635] The processing flow will be explained below.

[1636] Step 1:

[1637] The user enters the query "Kioicho" into the search box on the device and presses the search button.

[1638] Step 2:

[1639] The terminal sends the query "Kioicho" entered by the user to the server.

[1640] Step 3:

[1641] The server analyzes the received query, determines that it is an ambiguous query, and determines that it needs to use generative AI methods to generate an appropriate answer.

[1642] Step 4:

[1643] The server accesses the company's catalog database to retrieve the latest event information about Kioicho and related tourist spots, and also retrieves related information from the knowledge base.

[1644] Step 5:

[1645] The server's AI generator generates optimal answers based on the acquired data. These answers include a list of events currently being held in Kioicho and descriptions of recommended tourist spots.

[1646] Step 6:

[1647] The emotion engine recognizes and analyzes the user's emotions at that time based on their input, past behavioral data, facial expressions, tone of voice, etc. For example, it can determine emotions such as excitement, joy, or sadness based on the user's input speed and choice of words.

[1648] Step 7:

[1649] The server optimizes the response provided by the generative AI means based on the emotional information recognized by the emotion engine. For example, if the user is feeling stressed, the server will adjust the response to recommend relaxing tourist spots.

[1650] Step 8:

[1651] In addition to the generated answers, the server integrates map links to restaurants around Kioicho and related map information.

[1652] Step 9:

[1653] The server adjusts the display format of search results according to the user's emotions recognized by the emotion engine. For example, if the user feels like relaxing, the server displays the results in a calming interface.

[1654] Step 10:

[1655] The server generates the final search result data and sends it to the user's device. The search result data includes the generated answer, an event list, a list of tourist attractions, and a map link.

[1656] Step 11:

[1657] The device then displays the received search results to the user, such as a list of events currently being held in Kioicho, a list of tourist spots, and map links to nearby restaurants.

[1658] Step 12:

[1659] Users can check the search results and select events or spots they are interested in. If necessary, they can click on a map link to get more detailed information.

[1660] Example 2

[1661] 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."

[1662] Conventional search systems have difficulty providing appropriate answers to ambiguous queries entered by users, and they are unable to provide appropriate search results that reflect the user's emotions, resulting in a poor user experience.

[1663] 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.

[1664] In this invention, the server includes means for receiving an ambiguous query entered by a user, generation AI means for analyzing the ambiguous query and generating an appropriate answer, means for integrating the generated answer and related information to generate final search results, an emotion engine for recognizing and analyzing emotions from the user's input and past behavioral data, and means for optimizing the answer by the generation AI means based on the emotion information recognized by the emotion engine. This allows the user to quickly obtain appropriate search results even for an ambiguous query, and further provides an optimal answer according to the user's emotions, thereby improving the user experience.

[1665] A "query" is a word or phrase that a user enters to search for information.

[1666] A "generative AI means" is an artificial intelligence technology for analyzing a received query and generating an appropriate response to that query.

[1667] The "emotion engine" is a function that recognizes and analyzes emotions from user input and past behavioral data.

[1668] A "database" is a collection of specific information that is systematically organized, stored, and can be accessed as needed.

[1669] A "knowledge base" is a database that systematically organizes and accumulates specific specialized knowledge and information.

[1670] "Cross-reference" is a function or method that allows related information to be mutually referenced.

[1671] The "integration means" is a function for compiling the generated answers and related information into a single result and providing it to the user.

[1672] "Optimization" is the process of adjusting or improving to obtain the best possible condition or result for a particular purpose or condition.

[1673] "Search results" are a collection of answers and related information provided in response to a user-entered query.

[1674] This invention is a system that uses a generative AI means and an emotion engine to provide appropriate search results for ambiguous queries entered by users. The system includes a user terminal, a server, an emotion engine, and a communication means for linking these. Furthermore, the system uses a database and knowledge base maintained by a company to fine-tune the generative AI means.

[1675] System Overview

[1676] User terminal

[1677] A user terminal is a device through which a user enters a query and receives and displays search results. Specific hardware examples include PCs, smartphones, and tablets. Users enter queries and view results through this terminal.

[1678] server

[1679] The server is responsible for parsing queries, generating answers using generative AI means, recognizing and analyzing emotions using an emotion engine, and integrating and transmitting search results. The server is equipped with a high-performance processor and a large amount of memory, allowing for real-time data processing.

[1680] Generation AI means

[1681] Generative AI is used to analyze queries and generate appropriate answers. Specifically, it uses machine learning algorithms to analyze queries and generate optimal answers based on relevant information. This generative AI is fine-tuned using the company's database and knowledge base.

[1682] Emotion Engine

[1683] The emotion engine is an engine that recognizes and analyzes emotions from user input and past behavioral data. Specifically, it analyzes data such as input speed, word choice, and query frequency to determine whether the user is feeling excitement, joy, sadness, stress, or other emotions.

[1684] Specific examples

[1685] An example of the behavior when a user enters the query "Kioicho" into the search box of a terminal will be described.

[1686] 1. Receiving a query: The user enters the query "Kioicho" in the search box of the device and presses the search button. The device sends this query to the server.

[1687] 2. Query Analysis: The server analyzes the received query and identifies its ambiguity. It determines that "Kioicho" is a place name and is associated with a wide range of data, including information about nearby tourist attractions and events.

[1688] 3. Data Acquisition: The server acquires the latest event information and tourist spot data about Kioicho from the company's database and knowledge base.

[1689] 4. Answer generation: Based on the data acquired by the generation AI means, the optimal answer is generated, including descriptions of tourist spots such as "Kioicho Garden" and "Historical Museum," and event information such as "2023 Spring Festival" and "Weekend Music Live."

[1690] 5. Emotion recognition and analysis: The emotion engine recognizes that the user is likely feeling stressed based on their typing speed and choice of words.

[1691] 6. Answer optimization: The server optimizes the answers based on the emotional information recognized by the emotion engine, so that relaxing spots (e.g., "Kioicho Garden") are given priority.

[1692] 7. Integration and transmission of search results: The server integrates the generated answers with map links to restaurants around Kioicho and related map information to generate the final search results and transmit them to the user's device.

[1693] 8. Displaying search results: The device displays the received search results to the user. Specifically, it displays a list of events being held in Kioicho, Kioicho Garden, and a map link to nearby restaurants. The user can check this information and click to view more detailed information as needed.

[1694] In this way, users can quickly and accurately find the information they are looking for, even for ambiguous queries, making search behavior more efficient. Utilizing an emotion engine is expected to significantly improve the quality of the user experience.

[1695] Prompt Sentence Examples

[1696] As an example of a user entering a query, consider the following prompt:

[1697] "Please tell me about historical tourist spots in Kioicho."

[1698] "What events are happening in Kioicho?"

[1699] "Please tell me some recommended restaurants around Kioicho."

[1700] Searches are performed based on these prompts, and generative AI methods and an emotion engine work together to provide optimal search results.

[1701] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1702] Step 1:

[1703] Receiving a query

[1704] The user enters the query "Kioicho" into the search box on the device and presses the search button.

[1705] Input: User-entered query "Kioicho"

[1706] Output: Data sent from the device to the server for the query "Kioicho"

[1707] Specific operation: When the search button on the terminal interface is clicked, query data is constructed and sent to the server.

[1708] Step 2:

[1709] Parsing a query

[1710] The server analyzes the received query and identifies any ambiguities.

[1711] Input: Query "Kioicho" sent from the terminal

[1712] Output: Information about the ambiguity of the query (e.g., it is a place name and may be related to tourist attractions or event information)

[1713] What happens: The server uses natural language processing algorithms to analyze the meaning of the query and identify any ambiguities.

[1714] Step 3:

[1715] Data Acquisition

[1716] The server retrieves the latest event information and tourist spot data about Kioicho from the company's database and knowledge base.

[1717] Input: Query ambiguity information and related database lookup request

[1718] Output: Related event information and tourist spot data (e.g., "2023 Spring Festival," "Kioicho Garden," etc.)

[1719] Specific operation: The server calls the database API, obtains the information corresponding to the query, and stores it in the cache.

[1720] Step 4:

[1721] Generate answers

[1722] The generation AI means generates the optimal answer based on the data acquired.

[1723] Input: Event information and tourist spot data obtained from the database

[1724] Output: A specific answer to be provided to the user (e.g., "Kioicho Garden is a beautiful garden and a great place to relax," "There is live music on weekends," etc.)

[1725] Specific operation: The generative AI means converts the acquired data using a natural language generation algorithm to make it easier for humans to understand.

[1726] Step 5:

[1727] Emotion Recognition and Analysis

[1728] The emotion engine recognizes and analyzes the user's emotions at that time based on their input speed and choice of words.

[1729] Input: User typing speed, words used, and past behavior data

[1730] Output: Information about the user's emotional state (e.g., feeling stressed)

[1731] How it works: The emotion engine uses machine learning algorithms to determine emotions from input data.

[1732] Step 6:

[1733] Optimizing answers

[1734] The server optimizes the answer provided by the generative AI means based on the emotional information recognized by the emotion engine.

[1735] Input: Generated answers and information about the user's emotional state

[1736] Output: Optimized answers (e.g., prioritize relaxation spots for stressed users)

[1737] What happens: The server re-evaluates the answer and adjusts it accordingly.

[1738] Step 7:

[1739] Search result consolidation and submission

[1740] The server then integrates the generated answers with map links to restaurants around Kioicho and other related map information to generate the final search results, which are then sent to the user's device.

[1741] Input: Optimized answer and additional information (e.g. restaurant map link, map information)

[1742] Output: Final search result data

[1743] Specific operation: The server compiles information from various data sources, constructs the final search results, and sends them to the device.

[1744] Step 8:

[1745] Displaying search results

[1746] The terminal displays the received search results to the user.

[1747] Input: Final search result data sent from the server

[1748] Output: Search results displayed on the screen (e.g., "List of events being held in Kioicho," "Kioicho Garden," "Map links to nearby restaurants")

[1749] What happens: The device analyzes the received data and displays it appropriately in the user interface. The user can click on the information of interest to view further details.

[1750] (Application example 2)

[1751] 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."

[1752] Conventional search systems have difficulty providing satisfactory search results when users enter ambiguous queries. Providing optimal search results is particularly challenging when the information desired varies depending on the user's emotions. Food delivery services also face challenges in recommending appropriate restaurants and menus based on ambiguous queries and user emotions.

[1753] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving an ambiguous query entered by a user; a generation AI means for analyzing the ambiguous query and generating an appropriate answer; an emotion engine means for analyzing the user's emotions and optimizing the generated answer and related information; a means for integrating the generated answer and related information to generate final search results; and a means for transmitting the search results to the user's terminal and displaying them to the user. This makes it possible to provide optimal search results that take the user's emotions into consideration, even when the user enters an ambiguous query. Furthermore, in a food delivery service, optimal restaurants and menus can be recommended based on the ambiguous query, thereby improving user satisfaction.

[1754] An "ambiguous query" is an unclear search term that a user enters without providing any specific information.

[1755] "Generative AI methods" refers to artificial intelligence techniques for analyzing ambiguous queries and generating appropriate answers.

[1756] The "emotion engine" is an engine that analyzes emotions from user input and behavioral data and optimizes the generated information.

[1757] "User terminal" refers to the device used by the user to perform a search, including smartphones and personal computers.

[1758] A "corporate catalog" is a database that lists the products and services offered by a company.

[1759] A "knowledge base" is a database that systematically compiles the knowledge and information held by a company.

[1760] "Search Results" refers to answers and related information generated based on a user-entered query.

[1761] "Map information" is data that indicates geographical location information included in the search results.

[1762] "Route information" is data that provides guidance for a user to reach a particular location.

[1763] This invention is a system that provides optimal search results for ambiguous queries entered by users using generative AI means and an emotion engine. This system is embodied as an application called "FoodGenie" that is specialized for food delivery services.

[1764] Program processing

[1765] 1. Receiving a query:

[1766] When a user enters an ambiguous query (e.g., "pizza") on a smartphone app, the device sends this query to a server.

[1767] 2. Emotion analysis:

[1768] The server sends the query to the emotion engine, which analyzes the user's emotional state based on the user's input speed and word choice.

[1769] 3. Answer generation by generative AI:

[1770] The server accesses the company's catalog or knowledge base to retrieve the necessary data, and the generative AI method generates the appropriate answer based on this.

[1771] 4. Emotion-based optimization:

[1772] The generated answers are optimized based on the analysis results from the emotion engine. For example, if the user is feeling stressed, it will recommend a "relaxing family pizza set."

[1773] 5. Search Results Integration:

[1774] The server integrates the generated answers, along with any associated map or route information, to produce the final search results.

[1775] 6. Submitting and Displaying Search Results:

[1776] The server sends the final search results to the user's device and displays them to the user, who then displays the results in the app's interface.

[1777] Hardware and software used

[1778] Device: Smartphone (iOS, Android)

[1779] Emotion Engine API: Emotion-API Platform

[1780] Generate AI model: Custom AI Model API (e.g. OpenAI, GPT-4)

[1781] Server: Cloud server (AWS, Google Cloud)

[1782] Specific examples

[1783] For example, if a user types "pizza" into their smartphone, the emotion engine will detect "stress." The generative AI will prioritize recommendations for restaurants that offer "relaxing family pizza sets."

[1784] Example prompt for a generative AI model:

[1785] plaintext

[1786] A user types "pizza." The user's emotion is "stress." Can you recommend a restaurant and menu that will help them relax and relieve stress?

[1787] This system can provide optimal search results that take emotions into account, even when a user enters an ambiguous query.Furthermore, in food delivery services, it can recommend optimal restaurants and menus based on ambiguous queries, thereby improving user satisfaction.

[1788] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1789] Step 1:

[1790] A user enters a vague query into the "FoodGenie" app on their smartphone and presses the search button. The entered query is a non-specific phrase such as "pizza." Once the input is received, the device sends the query information to the server. The input data includes the query text and the user ID.

[1791] Step 2:

[1792] The server sends a request to the emotion engine API to analyze the received query. The emotion engine takes the query text and user ID as input and analyzes the user's emotional state. For example, it determines the user's emotion (e.g., stress or fatigue) from the query input speed and the words used. The emotion information is returned to the server.

[1793] Step 3:

[1794] The server accesses the company's catalog and knowledge base to gather data related to the query. For example, the server retrieves multiple restaurant and menu data related to "pizza." The results collected from the database include restaurant names, menus, ratings, location information, etc.

[1795] Step 4:

[1796] The server uses a generative AI to analyze the target data and generate the optimal answer. At this time, the server uses the collected restaurant data and the user's emotional information obtained from the emotion engine as input. For example, if it is determined that the user is feeling stressed, the generative AI will suggest a "relaxing family pizza set to relieve stress." The generated answer data includes information such as the restaurant name, recommended menu items, and ratings.

[1797] Step 5:

[1798] The server further integrates the generated answers and refines the final search results. Specifically, it adds map information and route information for the restaurant. It also optimizes the interface based on the analysis results of the emotion engine (for example, displaying calmer colors for users with high stress levels). The final search result data is generated.

[1799] Step 6:

[1800] The server then sends the final search results to the user's device, which then displays them on the app's interface, including the name of the restaurant the user is looking for, map information, and recommended menu items.

[1801] Step 7:

[1802] Users can review the search results displayed in the app and select the restaurant or menu they are interested in. This allows users to easily obtain information and directions to specific restaurants.

[1803] 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.

[1804] 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.

[1805] 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.

[1806] 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.

[1807] 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.

[1808] 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.

[1809] 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).

[1810] 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.

[1811] 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."

[1812] 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.

[1813] 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).

[1814] 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.

[1815] 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.

[1816] 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.

[1817] 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.

[1818] 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.

[1819] 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.

[1820] 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.

[1821] 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.

[1822] 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.

[1823] 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.

[1824] The following is further disclosed regarding the above embodiment.

[1825] (Claim 1)

[1826] means for receiving a user-entered ambiguous query;

[1827] a generative AI means for analyzing ambiguous queries and generating appropriate answers;

[1828] a means for integrating the generated answers and related information to produce a final search result;

[1829] means for transmitting the search results to a user's terminal and displaying them to the user;

[1830] A system including:

[1831] (Claim 2)

[1832] 10. The system of claim 1, further comprising means for retrieving data from a catalog or knowledge base maintained by the enterprise and using the data to fine-tune the generative AI means.

[1833] (Claim 3)

[1834] 10. The system of claim 1, further comprising means for including map information and associated search trends in the search results.

[1835] "Example 1"

[1836] (Claim 1)

[1837] means for receiving a user-entered ambiguous query;

[1838] a means for analyzing ambiguous queries and identifying their ambiguity;

[1839] a means of obtaining relevant data from databases maintained by the company;

[1840] A means for generating optimal answers using a generative AI model; and

[1841] a means for integrating the generated answers and associated map information to generate a final search result;

[1842] means for transmitting the final search results to the user's terminal and displaying them to the user;

[1843] A system including:

[1844] (Claim 2)

[1845] 10. The system of claim 1, further comprising means for retrieving data from a catalog or knowledge base maintained by an enterprise and using the data to fine-tune the generative AI model.

[1846] (Claim 3)

[1847] 10. The system of claim 1, further comprising means for including map information and data using APIs of related external services in the search results.

[1848] "Application Example 1"

[1849] (Claim 1)

[1850] means for receiving a user-entered ambiguous query;

[1851] a generative AI means for analyzing ambiguous queries and generating appropriate answers;

[1852] a means for integrating the generated answers and related information to produce a final search result;

[1853] means for transmitting the search results to a user's terminal and displaying them to the user;

[1854] A means for users to obtain search information through hardware located in a physical store;

[1855] A system including:

[1856] (Claim 2)

[1857] 10. The system of claim 1, further comprising means for retrieving data from a catalog or knowledge base maintained by the enterprise and using the data to fine-tune the generative AI means.

[1858] (Claim 3)

[1859] 10. The system of claim 1, further comprising means for including map information and associated search trends in the search results.

[1860] "Example 2: Combining Emotion Engines"

[1861] (Claim 1)

[1862] means for receiving a user-entered ambiguous query;

[1863] a generative AI means for analyzing ambiguous queries and generating appropriate answers;

[1864] a means for integrating the generated answers and related information to produce a final search result;

[1865] means for transmitting the search results to a user's terminal and displaying them to the user;

[1866] An emotion engine that recognizes and analyzes emotions from user input and past behavioral data,

[1867] A means for optimizing the response by the generating AI means based on the emotional information recognized by the emotion engine;

[1868] A system including:

[1869] (Claim 2)

[1870] 10. The system of claim 1, further comprising means for retrieving data from a database or knowledge base maintained by the enterprise and using the data to fine-tune the generative AI means.

[1871] (Claim 3)

[1872] 10. The system of claim 1, further comprising means for including map information and related information in the search results.

[1873] "Application example 2 when combining emotion engines"

[1874] (Claim 1)

[1875] means for receiving a user-entered ambiguous query;

[1876] a generative AI means for analyzing ambiguous queries and generating appropriate answers;

[1877] an emotion engine means for analyzing the emotion of the user and optimizing the generated answers and related information;

[1878] a means for integrating the generated answers and related information to produce a final search result;

[1879] means for transmitting the search results to a user's terminal and displaying them to the user;

[1880] A system including:

[1881] (Claim 2)

[1882] 10. The system of claim 1, further comprising means for retrieving data from a catalog or knowledge base maintained by an enterprise and utilizing the data to fine-tune the generative AI means.

[1883] (Claim 3)

[1884] 10. The system of claim 1, further comprising means for including map information and associated driving directions in the search results. [Explanation of symbols]

[1885] 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 user-entered ambiguous query; a generative AI means for analyzing ambiguous queries and generating appropriate answers; a means for integrating the generated answers and related information to produce a final search result; means for transmitting the search results to a user's terminal and displaying them to the user; A system including:

2. 10. The system of claim 1, further comprising means for obtaining data from a catalog or knowledge base maintained by an enterprise and using the data to fine-tune the generative AI means.

3. The system of claim 1 further comprising means for including map information and associated search trends in the search results.

Images