System

Abstract

A system is provided.SOLUTION: A system comprising: means for extracting a search query related to a disaster; means for generating a search intention for the search query; means for obtaining a search result based on the search query; means for comparing the search result with the search intention to determine whether the search result matches the search intention; and means for listing a search query that does not match the search intention.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] Currently, during disasters, many people search the Internet for important information, but there is no sufficient guarantee that their search queries will lead to relevant search results. This can prevent users from quickly and accurately obtaining the necessary information, putting lives and property at direct risk. Furthermore, it is difficult to respond to minority search queries, resulting in some needs not being met. Therefore, the objective of this invention is to provide a system that can provide appropriate information in response to search queries. [Means for solving the problem]

[0005] In order to solve the above-mentioned problems, the present invention provides the following means. First, a means for extracting search queries related to disasters is provided. Next, a means for generating search intent for the extracted search queries is provided. Also, a means for obtaining actual search results based on the search queries is provided, and a means for comparing the obtained search results with the generated search intent to determine whether they match. Furthermore, a means for listing search queries that do not match the search intent is provided. This system makes it possible to pick out search queries that provide search results that do not match the search intent, and to generate and provide appropriate answer modules based on the search results.

[0006] A "search query" is a string of characters or keywords that a user enters into an Internet search engine.

[0007] "Search intent" refers to the information or goal a user is seeking when they enter a particular search query.

[0008] A "search log" is historical data about queries entered by users into Internet search engines and the search results they produce.

[0009] "Answer Module" means a program or component that provides relevant information or content generated in response to a search query.

[0010] "Search results" refers to the list of web pages or information returned by a search engine based on a search query.

[0011] "Compare" refers to the process of evaluating how well search results match the search intent.

[0012] "Listing" refers to extracting search queries that do not match the search intent and organizing them in a list format.

[0013] "Publishing" means posting the generated answer module on a website or app so that it can be viewed by users. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0022] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0035] The present invention provides a system for quickly and accurately providing appropriate information based on a user's search query during a disaster. The system includes the following main components:

[0036] Query Extraction Phase

[0037] First, the server has the function of extracting search queries related to disasters. This is achieved by analyzing search engine search logs and filtering relevant queries. For example, queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage" are extracted and stored in a database.

[0038] Intention generation phase

[0039] Next, the server uses generation AI to generate search intent for the extracted search query. For example, for the query "nearby shelter," it generates intent such as "I'm looking for the specific location and access method of the shelter closest to my current location." This generated search intent is stored in a database.

[0040] Matching Phase

[0041] The server then automatically retrieves the actual search results for each query. This is done by using search engine APIs to retrieve the current search results for the query. For example, search for "nearby shelter" and collect the contents of the top search results. The server then compares the retrieved search results with the generated search intent to determine whether they match. If the search results do not match the user's search intent, the server lists the query.

[0042] List Output Phase

[0043] This list is used to notify administrators of queries that do not match. For example, queries such as "nearby evacuation shelters," "areas with water outages," and "emergency contact information" are listed, and the server provides this list to the device, prompting the person in charge or administrator to take action.

[0044] Answer module generation phase

[0045] Finally, the server generates a tailored response module based on the list, for example, content about current evacuation shelter updates or water outages, utilizing a geographic information system (GIS) and real-time data feeds. The response module is then posted on a website or app for easy access by users.

[0046] Specific examples

[0047] Here are some specific examples from areas where earthquakes have occurred.

[0048] Assumed scenario

[0049] The user searches for "nearby shelter." The server analyzes that similar queries are frequently searched by multiple users and extracts the query "nearby shelter." It then uses a generation AI to generate the search intent: "The user is looking for information on the nearest shelter that can be reached from their current location."

[0050] The server uses a search engine API to retrieve current search results for "nearby shelters," and if the results only contain local history and general shelter information, but lack real-time shelter location information, it determines that the query is not a match. It then lists the query and notifies the administrator.

[0051] The server generates a response module that provides the latest evacuation shelter location information and posts it on the website or app. This process allows the system to provide users with the information they need in the event of a disaster.

[0052] The processing flow will be explained below.

[0053] Step 1:

[0054] The server analyzes search logs related to disasters, using a technique to filter queries containing specific keywords from search engine access logs.

[0055] Step 2:

[0056] The server stores the extracted search queries in a database, including the queries themselves and their frequency of searches.

[0057] Step 3:

[0058] The server uses generative AI to generate search intent for each search query stored in the database, passing the query through an AI model that extracts from it what the user is looking for.

[0059] Step 4:

[0060] The server stores the generated search intent in a database, which identifies the specific purpose or intent associated with each query.

[0061] Step 5:

[0062] The server uses the search engine API to automatically retrieve the current search results for each query, including the content of the top-ranking pages from the search engine.

[0063] Step 6:

[0064] The server analyzes the content of the search results and compares them with the search intent stored in the database to determine whether the search results match the intent.

[0065] Step 7:

[0066] The server lists queries for which search results do not match the search intent. The list includes the queries that do not match, and this information is notified to the administrator.

[0067] Step 8:

[0068] Based on the list, the server generates a response module that matches the intent, including up-to-date information on evacuation shelters and water outages.

[0069] Step 9:

[0070] The server then posts the generated answer module on a website or app, allowing users to easily access it and obtain the information they need.

[0071] Example 1

[0072] 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."

[0073] During a disaster, it is important for users to be able to access the information they need quickly and accurately. However, with conventional systems, search results often do not match the user's search intent, making it difficult to quickly obtain the information they need. Furthermore, the lack of appropriate information can increase user confusion and anxiety, creating a problem. Another issue is that information is not updated in real time, even though many users are searching for similar queries.

[0074] 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.

[0075] In this invention, the server includes means for extracting search queries related to disasters, means for generating search intent for the search queries, means for analyzing whether the search queries are searched by many users, means for saving the generated search intent in a database, means for comparing search results with the generated search intent to determine whether they match, means for listing search queries that do not match the search intent, means for notifying an administrator of the list, means for generating an answer module based on the search intent, means for displaying the answer module, and means for generating the answer module using a real-time data feed or a geographic information system. This makes it possible to quickly and accurately provide information desired by users in the event of a disaster.

[0076] A "search query" is a word or phrase that a user enters into a search engine to find information.

[0077] "Search intent" refers to the information or goal a user is really looking for when they enter a search query.

[0078] "Search results" are the list of web pages or information that a search engine displays based on a user's search query.

[0079] A "database" is a system for efficiently storing, managing, and searching large amounts of data.

[0080] A "generative AI model" is an algorithm or program that uses artificial intelligence to analyze data and make predictions or generate results that suit a specific purpose.

[0081] A "real-time data feed" is a data stream that updates with ongoing information.

[0082] A "geographic information system" is a system for collecting, managing, analyzing, and displaying geographic data and spatial information.

[0083] "Notification means" refers to a method or system for notifying relevant parties of specific information.

[0084] The present invention provides a system for providing quick and accurate information based on a user's search query during a disaster. The system includes the following main components:

[0085] 1. Query extraction phase

[0086] The server extracts search queries relevant to disasters from search engine search logs. Specifically, it uses large-scale data processing tools such as Apache Hadoop and Google BigQuery to analyze the search logs and filter queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage." The extracted queries are stored in a database (for example, MySQL or PostgreSQL).

[0087] 2. Search Intent Generation Phase

[0088] Next, the server uses a generative AI model (e.g., OpenAI's GPT-4) to generate the search intent for the extracted search query. The server inputs the query into the generative AI model and generates the intent, "The user is looking for information about the nearest shelter from their current location." This generated search intent is stored in a database. An example of an input prompt sentence is, "Please generate the user's search intent based on the search query 'nearby shelters'."

[0089] 3. Matching Phase

[0090] The server uses a search engine API (e.g., Google Search API) to retrieve current search results for a query. For the query "shelters near me," the server collects the top search results and compares them with the generated search intent. If the search results do not provide real-time shelter information, the server determines that they do not match the intent. This information is stored in a database.

[0091] 4. List output phase

[0092] The server lists search queries that do not match the search intent and notifies the administrator of this list. Notifications are sent via email or notification services (e.g., Twilio or Slack API). The administrator's dashboard displays queries such as "nearby evacuation centers," "areas with water outages," and "emergency contact information."

[0093] 5. Answer module generation phase

[0094] The server generates answer modules based on the listed queries, leveraging geographic information systems (GIS) and real-time data feeds to generate content such as up-to-date evacuation shelter locations and water outage status. The answer modules are then posted on websites and apps for easy access by users.

[0095] Specific examples

[0096] Let's take the case of an earthquake as an example. When a user searches for the query "nearby shelter," the server first analyzes the search log and extracts this query. Next, it uses a generative AI model to generate the search intent: "The user is looking for information on the shelter closest to their current location." The server retrieves search results for this query using a search engine API, and if it determines that real-time shelter information is missing, it determines that the results are not a match. This information is listed and notified to the administrator. Then, it generates an answer module based on the latest shelter location information and posts it on a website or app, allowing users to quickly and accurately access the information they need. This example ensures the accuracy and speed of information during disasters.

[0097] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0098] Step 1:

[0099] Query Extraction Phase

[0100] Input: The server receives the search logs of a search engine.

[0101] How it works: The server uses Apache Hadoop and Google BigQuery to analyze search logs. Specifically, it periodically collects search logs and filters search queries that include disaster-related keywords such as "earthquake occurrence," "evacuation location," and "water outage."

[0102] Output: The filtered search query is saved in a database.

[0103] Step 2:

[0104] Search Intent Generation Phase

[0105] Input: The server reads the extracted search query from the database.

[0106] How it works: The server inputs a query to a generative AI model (e.g., OpenAI's GPT-4) using a prompt sentence. An example of a prompt sentence is "Generate the user's search intent based on the search query 'nearby shelters'." The generative AI model analyzes the search query and generates the search intent. It generates the intent, "The user is looking for information about the shelter closest to their current location."

[0107] Output: The generated search intent is stored in a database.

[0108] Step 3:

[0109] Matching Phase

[0110] Input: The server reads the generated search intent from the database.

[0111] How it works: The server uses a search engine API (e.g., Google Search API) to retrieve current search results for a search query. The content of the retrieved search results is analyzed and compared to the generated search intent. For example, if the query "near me shelters" returns search results that are mostly about local history and general shelter information, but lack real-time shelter location information, it is determined to be a poor match.

[0112] Output: The queries that were not found to be a match are listed and stored in a database.

[0113] Step 4:

[0114] List Output Phase

[0115] Input: The server reads the listed unmatched queries.

[0116] Behavior: The server notifies the administrator of this list by running a function that displays it on the administrator's dashboard, and also notifies the administrator via an email system or notification service (e.g., Twilio, Slack API).

[0117] Output: The administrator sees a list of unmatched queries such as "nearest shelter," "area without water," and "emergency contact information."

[0118] Step 5:

[0119] Answer module generation phase

[0120] Input: The server gathers the necessary information based on the listed queries.

[0121] How it works: The server uses real-time data feeds and geographic information systems (GIS) to obtain the latest information. For example, it collects data on the latest locations of evacuation centers and water outages. It then generates answer modules based on this data. It then uses generative AI models to automatically generate content to provide users with the information they are looking for.

[0122] Output: The generated answer module is posted on a website or app for easy access by users.

[0123] (Application example 1)

[0124] 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."

[0125] In the event of a disaster, rapid and accurate information provision is required, but conventional systems sometimes have difficulty providing information that matches the user's search intent. Furthermore, in business environments such as factories, there is a lack of means for personnel to instantly obtain and display the emergency information they need. This can lead to delayed responses in the event of a disaster, potentially hindering safety.

[0126] 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.

[0127] In this invention, the server includes means for extracting search queries relevant to disasters, means for generating search intent for the search queries, means for acquiring search results based on the search queries, means for comparing the search results with the search intent to determine whether they match, means for listing search queries that do not match the search intent, means for acquiring and displaying real-time data based on the listed search queries, and means for displaying the real-time data on a user terminal. This allows for the rapid and accurate provision of information required by users, enabling rapid emergency response, particularly in business environments such as factories.

[0128] "Disaster" refers to the environment or situation when an emergency occurs due to a natural or man-made disaster.

[0129] A "search query" refers to an inquiry or keyword that a user enters into a search system.

[0130] "Search intent" refers to the purpose or true intention of a user when entering a search query.

[0131] "Search results" refers to the set of information and links that a search system returns in response to a search query.

[0132] "Match" refers to a situation where the search intent matches the search results and meets the user's needs.

[0133] "Listing" refers to extracting elements based on specific conditions and compiling them into a list.

[0134] "Real-time data" refers to the latest information at the current time or data that is updated in real time.

[0135] "User terminal" refers to a device used by a user to access information, such as a smartphone, smart glasses, or head-mounted display.

[0136] An "answer module" refers to specific information delivery methods or content generated based on search intent.

[0137] "Posting" refers to making information publicly available on a display device or platform.

[0138] The present invention is a specific embodiment of a system for providing rapid and accurate information based on user search queries during disasters. This system consists of a server, a user terminal (such as smart glasses), and a generative AI model.

[0139] Server Processing

[0140] The server extracts search queries related to disasters from search logs and generates search intent based on them. It then obtains search results based on the search queries using a search engine API and compares the obtained search results with the generated search intent. Based on this comparison, it lists search queries that do not match the user's search intent.

[0141] The server is also responsible for retrieving real-time data based on the listed search queries and displaying it on the user's device. The real-time data can come from a GIS or real-time data feeds.

[0142] User terminal processing

[0143] The user terminal, specifically the smart glasses, displays the real-time data received from the server, allowing the user to take prompt and accurate evacuation and emergency response measures in the event of a disaster.

[0144] Hardware and software used

[0145] The server is built on a cloud computing platform (e.g., AWS, Google Cloud). The search engine API can be the Google Search API. Generative AI models such as OpenAI's GPT-3 and GPT-4 are used.

[0146] The user terminal uses smart glasses (e.g., Google Glass, Vuzix Blade), which allow the user to obtain visual information in real time.

[0147] Specific examples

[0148] Consider the case where an earthquake occurs inside a factory. When a worker searches for the query "nearby evacuation site," the server analyzes the search log and extracts this query. Using a generative AI model, it generates the search intent: "The user is looking for information on the nearest evacuation site that can be reached from their current location."

[0149] Next, the server uses a search engine API to retrieve search results for "nearby evacuation sites" and determines whether the displayed results match the user's intent. If not, it lists this query and retrieves the latest evacuation site information from real-time data. This information is displayed on the smart glasses, allowing personnel to take prompt evacuation action.

[0150] Example prompts for generative AI models

[0151] Explain why users might search for "shelters near me."

[0152] In this way, it becomes possible to provide real-time information based on the user's search intent, enabling a quick and accurate response in the event of a disaster.

[0153] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0154] Step 1:

[0155] The server extracts search queries related to disasters. Specifically, the server analyzes search logs and filters search queries containing related keywords such as "earthquake," "evacuation site," and "water outage." This allows specific queries used by users during disasters to be extracted. The input is search log data, and the output is the extracted search queries.

[0156] Step 2:

[0157] The server generates search intent for the extracted search queries. It uses a generative AI model (e.g., OpenAI's GPT-3) to generate user intent for each search query. The prompt sentence is "Please explain why users search for <query>." The input is the search query, and the output is the generated search intent.

[0158] Step 3:

[0159] The server retrieves search results based on the search query. It uses a search engine API (e.g., Google Search API) to find the current search results for each search query. The input is the search query, and the output is the search results.

[0160] Step 4:

[0161] The server compares the retrieved search results with the generated search intent to determine whether they match. Specifically, it compares the content of the search results with the generated intent, and if they do not match, it lists the queries. The input is the search results and the search intent, and the output is a list of queries that do not match.

[0162] Step 5:

[0163] The server retrieves real-time data based on the listed search queries, gathers up-to-date information from GIS and real-time data feeds, and generates specific answer modules for unmatched queries. The input is a list of unmatched queries, and the output is an answer module.

[0164] Step 6:

[0165] The server displays the generated response module on the user's device. Specifically, the generated real-time data is sent to and displayed on the user's device, such as smart glasses. This allows the user to quickly obtain the necessary disaster information. The response module is the input, and the real-time data displayed on the user's device is the output.

[0166] In this way, the server performs a series of processes, making it possible to quickly and accurately provide necessary information to user terminals in the event of a disaster.

[0167] 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.

[0168] This invention is a system for quickly and accurately providing appropriate information based on user search queries during disasters. This system incorporates an emotion engine that recognizes the user's emotions, making it possible to more accurately understand the user's search intent and provide appropriate information.

[0169] Components

[0170] The system of the present invention comprises the following main components:

[0171] Query Extraction Phase

[0172] First, the server has the function of extracting search queries related to disasters. This is achieved by analyzing search engine search logs and filtering relevant queries. For example, queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage" are extracted and stored in a database.

[0173] Intention generation phase

[0174] The server then uses generative AI to generate search intent for each search query stored in the database. For example, for the query "nearby shelter," the server generates intent such as "I'm looking for the specific location and access method of the shelter closest to my current location." This generated search intent is stored in the database.

[0175] Emotion Recognition Phase

[0176] Furthermore, as a unique feature of the present invention, the server utilizes an emotion engine to recognize the user's emotions. Based on the search query entered by the user, emotions such as anxiety, urgency, excitement, etc. are determined. This emotion information is incorporated into the generation of search intent.

[0177] Matching Phase

[0178] The server then automatically retrieves the actual search results for each query. This is done by using search engine APIs to retrieve the current search results for the query. For example, search for "nearby shelter" and collect the content of the top search results. The server then compares the retrieved search results with the generated search intent (including sentiment information) to determine whether they match. If the search results do not match the user's search intent, the server lists the query.

[0179] List Output Phase

[0180] This list is used to notify administrators of queries that do not match. For example, queries such as "nearby evacuation shelters," "areas with water outages," and "emergency contact information" are listed, and the server provides this list to the device, prompting the person in charge or administrator to take action.

[0181] Answer module generation phase

[0182] Finally, the server generates a tailored response module based on the list, for example, content about current evacuation shelter updates or water outages, utilizing a geographic information system (GIS) and real-time data feeds. The response module is then posted on a website or app for easy access by users.

[0183] Specific examples

[0184] Here are some specific examples from areas where earthquakes have occurred.

[0185] Assumed scenario

[0186] The user searches for "nearby shelter." The server analyzes that similar queries are frequently searched by multiple users and extracts the query "nearby shelter." It then uses a generation AI to generate the search intent: "The user is looking for information on the nearest shelter that can be reached from their current location."

[0187] At the same time, the server uses its emotion engine to recognize that the user is feeling extremely anxious, and accordingly modifies the search intent to read, "The user is in a great hurry and wants to know the exact location of the nearest evacuation shelter and how to access it quickly."

[0188] The server uses a search engine API to retrieve current search results for "nearby shelters," and if the results only contain local history and general shelter information, but lack real-time shelter location information, it determines that the query is not a match. It then lists the query and notifies the administrator.

[0189] The server generates a response module that provides the latest evacuation shelter location information and posts it on the website or app. This process allows the system to provide users with the information they need in the event of a disaster. The introduction of the emotion engine makes it possible to respond in accordance with the level of urgency and the user's emotional state, further improving the user's sense of security and convenience.

[0190] The processing flow will be explained below.

[0191] Step 1:

[0192] The server analyzes search logs related to disasters, using a technique to filter queries containing specific keywords from search engine access logs.

[0193] Step 2:

[0194] The server stores the extracted search queries in a database, including the queries themselves and their frequency of searches.

[0195] Step 3:

[0196] The server uses generative AI to generate search intent for each search query stored in the database, passing the query through an AI model that extracts from it what the user is looking for.

[0197] Step 4:

[0198] The server stores the generated search intent in a database, which identifies the specific purpose or intent associated with each query.

[0199] Step 5:

[0200] The server uses an emotion engine to recognize the user's emotions, such as anxiety, urgency, or excitement, based on the search query entered by the user.

[0201] Step 6:

[0202] The server adjusts the generated search intent based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling anxious, it determines that the user is looking for quicker and more specific information.

[0203] Step 7:

[0204] The server uses the search engine API to automatically retrieve the current search results for each query, including the content of the top-ranking pages from the search engine.

[0205] Step 8:

[0206] The server analyzes the content of the search results and compares them with the search intent (including emotional information) stored in the database. This comparison determines whether the search results match the intent.

[0207] Step 9:

[0208] The server lists queries for which search results do not match the search intent. The list includes the queries that do not match, and this information is notified to the administrator.

[0209] Step 10:

[0210] Based on the list, the server generates a response module that matches the intent, including up-to-date information on evacuation shelters and water outages.

[0211] Step 11:

[0212] The server then posts the generated answer module on a website or app, allowing users to easily access it and obtain the information they need.

[0213] Specific examples

[0214] We will explain specific examples from areas where earthquakes have occurred.

[0215] Assumed scenario

[0216] Step 1:

[0217] The user searches for "shelters near me."

[0218] Step 2:

[0219] The server analyzes that similar queries are being searched by multiple users and extracts the query "nearby shelter."

[0220] Step 3:

[0221] The server uses a generation AI to generate search intent such as, "The user is looking for information about the nearest evacuation shelter that can be reached from their current location."

[0222] Step 4:

[0223] Using its emotion engine, the server recognizes that the user is extremely anxious and modifies the search intent to read, "The user is in a great hurry and wants to know the exact location of the nearest evacuation shelter and how to access it quickly."

[0224] Step 5:

[0225] The server uses a search engine API to retrieve the current search results for "shelters near me."

[0226] Step 6:

[0227] The server collects the content of the displayed search results and determines that they contain local history and general shelter information, but do not contain real-time shelter location information.

[0228] Step 7:

[0229] The server lists queries that do not match the search intent and notifies the administrator.

[0230] Step 8:

[0231] The server generates an answer module that provides up-to-date shelter location information that matches the user's search intent.

[0232] Step 9:

[0233] The server posts the generated answer module on a website or app so that users can access it.

[0234] This process allows us to provide users with the information they need in the event of a disaster. The introduction of an emotion engine makes it possible to respond in accordance with the level of urgency and the user's emotional state, further improving the user's sense of security and convenience.

[0235] Example 2

[0236] 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."

[0237] Conventional information provision systems in the event of a disaster simply analyze the search intent of search queries without considering the user's emotions, making it difficult to provide information that reflects the level of urgency or psychological needs. Furthermore, search results often do not adequately match the user's search intent, making it necessary to provide information quickly and accurately.

[0238] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for extracting search queries related to disasters, means for generating search intent for the search queries, means for recognizing a user's emotion from the search queries, means for incorporating the emotion based on the search intent, means for acquiring search results based on the search queries, means for comparing the search results with the search intent and the emotion to determine whether they match, and means for listing search queries that do not match the search intent. This makes it possible to generate search intent taking the user's emotion into consideration, and to provide quick and accurate information that corresponds to the level of urgency and psychological needs.

[0239] A "search query" is a word or phrase that a user enters into a search engine to request specific information.

[0240] "Search intent" refers to the information or goal a user is actually seeking through a search query.

[0241] "Emotion" refers to the psychological state or feeling a user has when entering a search query.

[0242] A "server" is a computer system that provides services to other computers and terminals on a network.

[0243] A "means" is a method, device, or process for achieving a specific purpose.

[0244] "Search results" are the lists of information and links that a search engine returns in response to a user's search query.

[0245] "Listing" means selecting items that meet specific conditions as a list.

[0246] An "emotion engine" is software or algorithms that analyze and recognize emotions from a user's text or voice.

[0247] A "generative AI model" is an artificial intelligence (AI) system that is trained on large amounts of data to generate text, infer intent, recognize emotions, and more.

[0248] An "answer module" is a program or component that provides specific answer information generated based on the user's search intent and emotions.

[0249] This invention is a system for quickly and accurately providing appropriate information based on user search queries during disasters. This system incorporates an emotion engine that recognizes the user's emotions, enabling it to more accurately understand the user's search intent and provide appropriate information.

[0250] Hardware and software used

[0251] This system mainly uses the following hardware and software:

[0252] Server: A computer system for processing information.

[0253] Search engine API: An interface for retrieving search results (e.g., Google Search API, Bing Search API).

[0254] Generative AI models: AI models for text generation, intent inference, and emotion recognition (e.g., GPT-3).

[0255] Emotion engine: Software for analyzing user emotions (e.g., IBM Watson Tone Analyzer).

[0256] Database: A system for storing extracted search queries and generated intents.

[0257] NLP libraries: Natural language processing libraries (e.g., NLTK, spaCy) for analyzing search logs and processing text.

[0258] Specific operation of the system

[0259] 1. Query Extraction

[0260] The server periodically collects search engine log data. For example, it downloads the log data using a Python script and uses an NLP library to extract search queries related to disasters, such as "earthquake occurrence," "evacuation location," and "water outage."

[0261] The extracted queries are stored in a database.

[0262] 2. Generating Search Intent

[0263] The server generates search intent for the saved search query using a generative AI model (e.g., GPT-3). For example, it inputs a prompt such as "Please explain the user's intent for this search query." and stores the generated intent in a database.

[0264] 3. Emotion recognition

[0265] The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to determine emotions based on the search query entered by the user. For example, if a user searches for "nearby shelter," the emotion engine can recognize anxiety and urgency and add emotional information to the search intent.

[0266] 4. Obtain and compare search results

[0267] The server uses a search engine API (e.g., Google Search API) to obtain the current search results for the query.

[0268] The retrieved search results are compared with the generated search intent and sentiment information to determine whether they match.

[0269] 5. Listing

[0270] The server lists queries that do not match the search intent and notifies the administrator of this list, for example by outputting the mismatched queries to a CSV file and sending it by email.

[0271] 6. Generating the Answer Module

[0272] The server generates an answer module that reflects the search intent and sentiment based on the listed queries. It inputs a prompt message to the generative AI model saying, "Please generate a module that provides guidance based on the latest evacuation shelter information," and displays the generated module on a website or app.

[0273] Example scenario

[0274] For example, in an area where an earthquake has occurred, a user searches for "nearby shelter." The server analyzes search logs and extracts the query "nearby shelter." Using a generative AI model, it generates the search intent: "The user is looking for information on the nearest shelter from their current location." At the same time, it uses an emotion engine to recognize the user's anxiety and revise the search intent to "The user is in a hurry and wants to know the specific location of the nearest shelter and how to quickly access it." The server uses a search engine API to obtain current search results for "nearby shelter." If the results are inappropriate, it lists the query and notifies the administrator. Finally, the server generates an answer module that provides the latest shelter location information and posts it on a website or app, allowing users to quickly access it.

[0275] Prompt Sentence Examples

[0276] "Please explain the user's intent for this search query."

[0277] "Determine the sentiment based on the user's input query."

[0278] "Please generate a module that provides guidance based on the latest evacuation shelter information."

[0279] This system can quickly and accurately provide users with the information they need in the event of a disaster. The introduction of an emotion engine makes it possible to respond in accordance with the level of urgency and the user's emotional state, further improving user comfort and convenience.

[0280] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0281] Step 1: Query extraction phase

[0282] Input: Search engine log data

[0283] Processing: The server analyzes the search engine log data and extracts queries related to disasters. Specifically, it downloads the log data using a Python script and uses an NLP library (e.g., NLTK, spaCy) to filter queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage."

[0284] Output: A list of disaster-related search queries

[0285] Specific behavior:

[0286] The server periodically runs a Python script to retrieve log data.

[0287] The server uses an NLP library to extract queries containing keywords.

[0288] The extracted queries are stored in a database.

[0289] Step 2: Search intent generation phase

[0290] Input: A list of extracted search queries

[0291] Processing: The server uses a generative AI model (e.g., GPT-3) to generate search intent for the search query. It provides the generative AI model with a prompt, such as "Please explain the user's intent for this search query."

[0292] Output: Generated search intent

[0293] Specific behavior:

[0294] The server inputs each search query into a generative AI model.

[0295] The intent obtained from the generative AI model is stored in a database.

[0296] Step 3: Emotion Recognition Phase

[0297] Input: Extracted search query

[0298] Processing: The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. Based on the search query entered by the user, it determines emotions such as anxiety or urgency.

[0299] Output: Emotional information (score)

[0300] Specific behavior:

[0301] The server sends the search query to the emotion engine API.

[0302] The idea is to incorporate sentiment scores obtained from the sentiment engine into search intent.

[0303] Step 4: Matching Phase

[0304] Input: Search intent and sentiment information

[0305] Processing: The server uses a search engine API (e.g., Google Search API) to retrieve search results and compare them with the generated search intent and sentiment information to determine if there is a match.

[0306] Output: A list of unmatched search queries

[0307] Specific behavior:

[0308] The server sends the search query to the search engine API and retrieves the results.

[0309] The search results are analyzed and compared with search intent and sentiment information.

[0310] As a result of the comparison, queries that are determined not to match are listed.

[0311] Step 5: List output phase

[0312] Input: A list of unmatched search queries

[0313] Action: The server notifies the administrator of unmatched queries, exports the list as a CSV file, emails it, or displays it in the web dashboard.

[0314] Output: Notification to administrator (CSV file or dashboard)

[0315] Specific behavior:

[0316] The server exports unmatched queries to a CSV file.

[0317] You can either email the administrator or view the list in the web dashboard.

[0318] Step 6: Answer module generation phase

[0319] Input: Unmatched search query and search intent and sentiment information

[0320] Processing: The server generates an answer module based on the unmatched query. For example, it gives a prompt such as "Please generate a module that provides guidance based on the latest evacuation shelter information" to the generative AI model to create appropriate content.

[0321] Output: Answer module

[0322] Specific behavior:

[0323] The server inputs a prompt sentence into the generative AI model and generates an answer module.

[0324] Deploy the generated answer module to your website or app.

[0325] In this way, information based on a user's search query during a disaster can be provided quickly and accurately according to the level of urgency and the user's emotional state.

[0326] (Application example 2)

[0327] 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."

[0328] In the event of a disaster, logistics center employees need to be able to obtain information quickly and appropriately so they can continue their work safely. However, existing systems lack the ability to provide real-time inventory management and delivery information that is appropriate for the unique circumstances of a disaster, and there are also insufficient means to alleviate employee anxiety. This creates problems that make it difficult to operate efficiently.

[0329] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0330] In this invention, the server includes means for extracting search queries relevant to disasters, means for generating search intent for the search queries, means for obtaining search results based on the search queries, means for comparing the search results with the search intent to determine whether they match, means for listing search queries that do not match the search intent, means for recognizing a user's emotion regarding the search queries, means for modifying the search intent based on the emotion, means for providing inventory management and delivery information in the event of a disaster, and means for providing information access using a terminal. This enables employees at a logistics center to obtain necessary information in real time and continue their work efficiently and safely even in the event of a disaster.

[0331] A "search query" is a keyword or phrase that a user enters when searching for information.

[0332] "Search intent" is the specific information or goal a user is seeking when they enter a search query.

[0333] A "search result" is a list of relevant information returned by a search engine in response to a search query.

[0334] "Emotion" refers to the psychological state or feeling a user has when entering a search query.

[0335] "Inventory management" is the activity of understanding the quantity and condition of goods held by logistics centers and companies and managing them appropriately.

[0336] "Shipping information" refers to detailed data and information regarding the shipping of products.

[0337] A "terminal" is a device that is connected to a computer system and inputs and outputs information, such as a smartphone or a head-mounted display.

[0338] MODE FOR CARRYING OUT THE INVENTION

[0339] The present invention is a system that enables employees at a logistics center to quickly and appropriately obtain information and continue their work safely in the event of a disaster, and is composed of the following main components:

[0340] Query Extraction Phase

[0341] First, the server has a means for extracting search queries relevant to disasters. This means analyzes search queries entered by logistics center employees using smartphones or head-mounted displays, and extracts queries that are particularly important in the event of a disaster, such as "delivery interruption," "stock shortage," and "emergency route." These queries are stored in a database.

[0342] Intention generation phase

[0343] The server then uses the generative AI model to generate search intent for each search query stored in the database. For example, for the query "delivery interruption," the server generates the search intent "I'm looking for safe delivery routes and relay points in the event of a disaster." This generated search intent is stored in the database.

[0344] Emotion Recognition Phase

[0345] Furthermore, the server has a means of recognizing employees' emotions using an emotion engine. Based on the search query entered by the user, emotions such as anxiety or urgency are determined and incorporated into the generation of search intent. This allows the search intent to be modified to be more specific and in line with the urgency, for example, "The user wants to know the evacuation route quickly."

[0346] Matching Phase

[0347] Next, the server retrieves the actual search results for each query using the search engine API. It compares the retrieved search results with the generated search intent (including sentiment information) to determine whether they match. For example, if only old route information is displayed for the query "emergency route," it is determined that the query does not match and the query is listed.

[0348] List Output Phase

[0349] The server notifies the administrator of the list of unmatched queries, allowing the administrator to take prompt action, such as providing updated delivery route information or updating inventory status.

[0350] Answer module generation phase

[0351] Finally, the server generates a response module based on the list, providing up-to-date evacuation route information and inventory status, utilizing GIS and real-time data feeds. The response module is displayed on the smartphones and head-mounted displays of logistics center employees for immediate access.

[0352] Specific examples

[0353] Usage Scenarios

[0354] For example, if a logistics center employee types the query "delivery interruption" into their smartphone, the server analyzes the query and generates the search intent: "I'm looking for safe delivery routes and relay points in the event of a disaster." At the same time, the server uses its emotion engine to recognize that the employee is feeling extremely anxious and modifies the search intent to "I want to know about evacuation routes quickly." If the results obtained by the server using the search engine API contain only outdated information, it determines that there is no match and notifies the administrator. An answer module providing the latest delivery route information is then generated and displayed on the employee's device. This process allows logistics center employees to quickly obtain the information they need, even in the event of a disaster, and continue their work efficiently and safely.

[0355] Example of input prompt for generative AI model

[0356] For the query "delivery disruption," generate search intent that reflects a distribution center employee wanting the best delivery route during a disaster.

[0357] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0358] Step 1:

[0359] The server extracts disaster-related search queries by analyzing search logs and filtering queries containing specific keywords (e.g., "delivery interruption," "stock shortage," "emergency route," etc.). The input is the search logs, and the output is a list of disaster-related search queries.

[0360] Step 2:

[0361] The server uses a generative AI model to generate search intent for each extracted search query. For example, for the query "delivery interruption," the server generates the search intent "I'm looking for safe delivery routes and relay points in the event of a disaster." The input is the extracted search query, and the output is the corresponding search intent.

[0362] Step 3:

[0363] The server uses an emotion engine to recognize the emotion of the user (employee) who entered the search query. In this phase, emotions such as anxiety or urgency are determined based on the entered keywords and phrases, and information is added. The input is the search query, and the output is the search intent, including emotional characteristics.

[0364] Step 4:

[0365] The server uses the search engine API to get the current search results based on the revised search intent. Specifically, it calls the search engine API to get the "delivery interruption" information in real time. The input of this step is the revised search intent, and the output is a list of relevant search results.

[0366] Step 5:

[0367] The server compares the retrieved search results with the generated search intent and determines whether they match. For example, it checks whether the search results contain information about safe routes or intermediate stops. The input is the search results and the search intent, and the output is the result of whether they match.

[0368] Step 6:

[0369] The server lists queries that do not match the search intent and notifies the administrator, for example, when up-to-date delivery route information is missing. The input is the unmatched query, and the output is a notification message.

[0370] Step 7:

[0371] The server generates response modules based on the list, providing up-to-date evacuation route information and inventory availability, leveraging a geographic information system (GIS) and real-time data feeds. The input is the latest information or data feed, and the output is the response module.

[0372] Step 8:

[0373] The generated response module is displayed on the user's smartphone or head-mounted display, allowing employees to quickly obtain the information they need in the event of a disaster. The input is the response module, and the output is the information displayed on the device.

[0374] This will enable logistics center employees to continue working efficiently and safely even in the event of a disaster.

[0375] 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.

[0376] 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.

[0377] 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.

[0378] [Second embodiment]

[0379] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0380] 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.

[0381] 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).

[0382] 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.

[0383] 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.

[0384] 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).

[0385] 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.

[0386] 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.

[0387] 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.

[0388] 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.

[0389] 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.

[0390] 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."

[0391] The present invention provides a system for quickly and accurately providing appropriate information based on a user's search query during a disaster. The system includes the following main components:

[0392] Query Extraction Phase

[0393] First, the server has the function of extracting search queries related to disasters. This is achieved by analyzing search engine search logs and filtering relevant queries. For example, queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage" are extracted and stored in a database.

[0394] Intention generation phase

[0395] Next, the server uses generation AI to generate search intent for the extracted search query. For example, for the query "nearby shelter," it generates intent such as "I'm looking for the specific location and access method of the shelter closest to my current location." This generated search intent is stored in a database.

[0396] Matching Phase

[0397] The server then automatically retrieves the actual search results for each query. This is done by using search engine APIs to retrieve the current search results for the query. For example, search for "nearby shelter" and collect the contents of the top search results. The server then compares the retrieved search results with the generated search intent to determine whether they match. If the search results do not match the user's search intent, the server lists the query.

[0398] List Output Phase

[0399] This list is used to notify administrators of queries that do not match. For example, queries such as "nearby evacuation shelters," "areas with water outages," and "emergency contact information" are listed, and the server provides this list to the device, prompting the person in charge or administrator to take action.

[0400] Answer module generation phase

[0401] Finally, the server generates a tailored response module based on the list, for example, content about current evacuation shelter updates or water outages, utilizing a geographic information system (GIS) and real-time data feeds. The response module is then posted on a website or app for easy access by users.

[0402] Specific examples

[0403] Here are some specific examples from areas where earthquakes have occurred.

[0404] Assumed scenario

[0405] The user searches for "nearby shelter." The server analyzes that similar queries are frequently searched by multiple users and extracts the query "nearby shelter." It then uses a generation AI to generate the search intent: "The user is looking for information on the nearest shelter that can be reached from their current location."

[0406] The server uses a search engine API to retrieve current search results for "nearby shelters," and if the results only contain local history and general shelter information, but lack real-time shelter location information, it determines that the query is not a match. It then lists the query and notifies the administrator.

[0407] The server generates a response module that provides the latest evacuation shelter location information and posts it on the website or app. This process allows the system to provide users with the information they need in the event of a disaster.

[0408] The processing flow will be explained below.

[0409] Step 1:

[0410] The server analyzes search logs related to disasters, using a technique to filter queries containing specific keywords from search engine access logs.

[0411] Step 2:

[0412] The server stores the extracted search queries in a database, including the queries themselves and their frequency of searches.

[0413] Step 3:

[0414] The server uses generative AI to generate search intent for each search query stored in the database, passing the query through an AI model that extracts from it what the user is looking for.

[0415] Step 4:

[0416] The server stores the generated search intent in a database, which identifies the specific purpose or intent associated with each query.

[0417] Step 5:

[0418] The server uses the search engine API to automatically retrieve the current search results for each query, including the content of the top-ranking pages from the search engine.

[0419] Step 6:

[0420] The server analyzes the content of the search results and compares them with the search intent stored in the database to determine whether the search results match the intent.

[0421] Step 7:

[0422] The server lists queries for which search results do not match the search intent. The list includes the queries that do not match, and this information is notified to the administrator.

[0423] Step 8:

[0424] Based on the list, the server generates a response module that matches the intent, including up-to-date information on evacuation shelters and water outages.

[0425] Step 9:

[0426] The server then posts the generated answer module on a website or app, allowing users to easily access it and obtain the information they need.

[0427] Example 1

[0428] 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."

[0429] During a disaster, it is important for users to be able to access the information they need quickly and accurately. However, with conventional systems, search results often do not match the user's search intent, making it difficult to quickly obtain the information they need. Furthermore, the lack of appropriate information can increase user confusion and anxiety, creating a problem. Another issue is that information is not updated in real time, even though many users are searching for similar queries.

[0430] 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.

[0431] In this invention, the server includes means for extracting search queries related to disasters, means for generating search intent for the search queries, means for analyzing whether the search queries are searched by many users, means for saving the generated search intent in a database, means for comparing search results with the generated search intent to determine whether they match, means for listing search queries that do not match the search intent, means for notifying an administrator of the list, means for generating an answer module based on the search intent, means for displaying the answer module, and means for generating the answer module using a real-time data feed or a geographic information system. This makes it possible to quickly and accurately provide information desired by users in the event of a disaster.

[0432] A "search query" is a word or phrase that a user enters into a search engine to find information.

[0433] "Search intent" refers to the information or goal a user is really looking for when they enter a search query.

[0434] "Search results" are the list of web pages or information that a search engine displays based on a user's search query.

[0435] A "database" is a system for efficiently storing, managing, and searching large amounts of data.

[0436] A "generative AI model" is an algorithm or program that uses artificial intelligence to analyze data and make predictions or generate results that suit a specific purpose.

[0437] A "real-time data feed" is a data stream that updates with ongoing information.

[0438] A "geographic information system" is a system for collecting, managing, analyzing, and displaying geographic data and spatial information.

[0439] "Notification means" refers to a method or system for notifying relevant parties of specific information.

[0440] The present invention provides a system for providing quick and accurate information based on a user's search query during a disaster. The system includes the following main components:

[0441] 1. Query extraction phase

[0442] The server extracts search queries relevant to disasters from search engine search logs. Specifically, it uses large-scale data processing tools such as Apache Hadoop and Google BigQuery to analyze the search logs and filter queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage." The extracted queries are stored in a database (for example, MySQL or PostgreSQL).

[0443] 2. Search Intent Generation Phase

[0444] Next, the server uses a generative AI model (e.g., OpenAI's GPT-4) to generate the search intent for the extracted search query. The server inputs the query into the generative AI model and generates the intent, "The user is looking for information about the nearest shelter from their current location." This generated search intent is stored in a database. An example of an input prompt sentence is, "Please generate the user's search intent based on the search query 'nearby shelters'."

[0445] 3. Matching Phase

[0446] The server uses a search engine API (e.g., Google Search API) to retrieve current search results for a query. For the query "shelters near me," the server collects the top search results and compares them with the generated search intent. If the search results do not provide real-time shelter information, the server determines that they do not match the intent. This information is stored in a database.

[0447] 4. List output phase

[0448] The server lists search queries that do not match the search intent and notifies the administrator of this list. Notifications are sent via email or notification services (e.g., Twilio or Slack API). The administrator's dashboard displays queries such as "nearby evacuation centers," "areas with water outages," and "emergency contact information."

[0449] 5. Answer module generation phase

[0450] The server generates answer modules based on the listed queries, leveraging geographic information systems (GIS) and real-time data feeds to generate content such as up-to-date evacuation shelter locations and water outage status. The answer modules are then posted on websites and apps for easy access by users.

[0451] Specific examples

[0452] Let's take the case of an earthquake as an example. When a user searches for the query "nearby shelter," the server first analyzes the search log and extracts this query. Next, it uses a generative AI model to generate the search intent: "The user is looking for information on the shelter closest to their current location." The server retrieves search results for this query using a search engine API, and if it determines that real-time shelter information is missing, it determines that the results are not a match. This information is listed and notified to the administrator. Then, it generates an answer module based on the latest shelter location information and posts it on a website or app, allowing users to quickly and accurately access the information they need. This example ensures the accuracy and speed of information during disasters.

[0453] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0454] Step 1:

[0455] Query Extraction Phase

[0456] Input: The server receives the search logs of a search engine.

[0457] How it works: The server uses Apache Hadoop and Google BigQuery to analyze search logs. Specifically, it periodically collects search logs and filters search queries that include disaster-related keywords such as "earthquake occurrence," "evacuation location," and "water outage."

[0458] Output: The filtered search query is saved in a database.

[0459] Step 2:

[0460] Search Intent Generation Phase

[0461] Input: The server reads the extracted search query from the database.

[0462] How it works: The server inputs a query to a generative AI model (e.g., OpenAI's GPT-4) using a prompt sentence. An example of a prompt sentence is "Generate the user's search intent based on the search query 'nearby shelters'." The generative AI model analyzes the search query and generates the search intent. It generates the intent, "The user is looking for information about the shelter closest to their current location."

[0463] Output: The generated search intent is stored in a database.

[0464] Step 3:

[0465] Matching Phase

[0466] Input: The server reads the generated search intent from the database.

[0467] How it works: The server uses a search engine API (e.g., Google Search API) to retrieve current search results for a search query. The content of the retrieved search results is analyzed and compared to the generated search intent. For example, if the query "near me shelters" returns search results that are mostly about local history and general shelter information, but lack real-time shelter location information, it is determined to be a poor match.

[0468] Output: The queries that were not found to be a match are listed and stored in a database.

[0469] Step 4:

[0470] List Output Phase

[0471] Input: The server reads the listed unmatched queries.

[0472] Behavior: The server notifies the administrator of this list by running a function that displays it on the administrator's dashboard, and also notifies the administrator via an email system or notification service (e.g., Twilio, Slack API).

[0473] Output: The administrator sees a list of unmatched queries such as "nearest shelter," "area without water," and "emergency contact information."

[0474] Step 5:

[0475] Answer module generation phase

[0476] Input: The server gathers the necessary information based on the listed queries.

[0477] How it works: The server uses real-time data feeds and geographic information systems (GIS) to obtain the latest information. For example, it collects data on the latest locations of evacuation centers and water outages. It then generates answer modules based on this data. It then uses generative AI models to automatically generate content to provide users with the information they are looking for.

[0478] Output: The generated answer module is posted on a website or app for easy access by users.

[0479] (Application example 1)

[0480] 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."

[0481] In the event of a disaster, rapid and accurate information provision is required, but conventional systems sometimes have difficulty providing information that matches the user's search intent. Furthermore, in business environments such as factories, there is a lack of means for personnel to instantly obtain and display the emergency information they need. This can lead to delayed responses in the event of a disaster, potentially hindering safety.

[0482] 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.

[0483] In this invention, the server includes means for extracting search queries relevant to disasters, means for generating search intent for the search queries, means for acquiring search results based on the search queries, means for comparing the search results with the search intent to determine whether they match, means for listing search queries that do not match the search intent, means for acquiring and displaying real-time data based on the listed search queries, and means for displaying the real-time data on a user terminal. This allows for the rapid and accurate provision of information required by users, enabling rapid emergency response, particularly in business environments such as factories.

[0484] "Disaster" refers to the environment or situation when an emergency occurs due to a natural or man-made disaster.

[0485] A "search query" refers to an inquiry or keyword that a user enters into a search system.

[0486] "Search intent" refers to the purpose or true intention of a user when entering a search query.

[0487] "Search results" refers to the set of information and links that a search system returns in response to a search query.

[0488] "Match" refers to a situation where the search intent matches the search results and meets the user's needs.

[0489] "Listing" refers to extracting elements based on specific conditions and compiling them into a list.

[0490] "Real-time data" refers to the latest information at the current time or data that is updated in real time.

[0491] "User terminal" refers to a device used by a user to access information, such as a smartphone, smart glasses, or head-mounted display.

[0492] An "answer module" refers to specific information delivery methods or content generated based on search intent.

[0493] "Posting" refers to making information publicly available on a display device or platform.

[0494] The present invention is a specific embodiment of a system for providing rapid and accurate information based on user search queries during disasters. This system consists of a server, a user terminal (such as smart glasses), and a generative AI model.

[0495] Server Processing

[0496] The server extracts search queries related to disasters from search logs and generates search intent based on them. It then obtains search results based on the search queries using a search engine API and compares the obtained search results with the generated search intent. Based on this comparison, it lists search queries that do not match the user's search intent.

[0497] The server is also responsible for retrieving real-time data based on the listed search queries and displaying it on the user's device. The real-time data can come from a GIS or real-time data feeds.

[0498] User terminal processing

[0499] The user terminal, specifically the smart glasses, displays the real-time data received from the server, allowing the user to take prompt and accurate evacuation and emergency response measures in the event of a disaster.

[0500] Hardware and software used

[0501] The server is built on a cloud computing platform (e.g., AWS, Google Cloud). The search engine API can be the Google Search API. Generative AI models such as OpenAI's GPT-3 and GPT-4 are used.

[0502] The user terminal uses smart glasses (e.g., Google Glass, Vuzix Blade), which allow the user to obtain visual information in real time.

[0503] Specific examples

[0504] Consider the case where an earthquake occurs inside a factory. When a worker searches for the query "nearby evacuation site," the server analyzes the search log and extracts this query. Using a generative AI model, it generates the search intent: "The user is looking for information on the nearest evacuation site that can be reached from their current location."

[0505] Next, the server uses a search engine API to retrieve search results for "nearby evacuation sites" and determines whether the displayed results match the user's intent. If not, it lists this query and retrieves the latest evacuation site information from real-time data. This information is displayed on the smart glasses, allowing personnel to take prompt evacuation action.

[0506] Example prompts for generative AI models

[0507] Explain why users might search for "shelters near me."

[0508] In this way, it becomes possible to provide real-time information based on the user's search intent, enabling a quick and accurate response in the event of a disaster.

[0509] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0510] Step 1:

[0511] The server extracts search queries related to disasters. Specifically, the server analyzes search logs and filters search queries containing related keywords such as "earthquake," "evacuation site," and "water outage." This allows specific queries used by users during disasters to be extracted. The input is search log data, and the output is the extracted search queries.

[0512] Step 2:

[0513] The server generates search intent for the extracted search queries. It uses a generative AI model (e.g., OpenAI's GPT-3) to generate user intent for each search query. The prompt sentence is "Please explain why users search for <query>." The input is the search query, and the output is the generated search intent.

[0514] Step 3:

[0515] The server retrieves search results based on the search query. It uses a search engine API (e.g., Google Search API) to find the current search results for each search query. The input is the search query, and the output is the search results.

[0516] Step 4:

[0517] The server compares the retrieved search results with the generated search intent to determine whether they match. Specifically, it compares the content of the search results with the generated intent, and if they do not match, it lists the queries. The input is the search results and the search intent, and the output is a list of queries that do not match.

[0518] Step 5:

[0519] The server retrieves real-time data based on the listed search queries, gathers up-to-date information from GIS and real-time data feeds, and generates specific answer modules for unmatched queries. The input is a list of unmatched queries, and the output is an answer module.

[0520] Step 6:

[0521] The server displays the generated response module on the user's device. Specifically, the generated real-time data is sent to and displayed on the user's device, such as smart glasses. This allows the user to quickly obtain the necessary disaster information. The response module is the input, and the real-time data displayed on the user's device is the output.

[0522] In this way, the server performs a series of processes, making it possible to quickly and accurately provide necessary information to user terminals in the event of a disaster.

[0523] 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.

[0524] This invention is a system for quickly and accurately providing appropriate information based on user search queries during disasters. This system incorporates an emotion engine that recognizes the user's emotions, making it possible to more accurately understand the user's search intent and provide appropriate information.

[0525] Components

[0526] The system of the present invention comprises the following main components:

[0527] Query Extraction Phase

[0528] First, the server has the function of extracting search queries related to disasters. This is achieved by analyzing search engine search logs and filtering relevant queries. For example, queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage" are extracted and stored in a database.

[0529] Intention generation phase

[0530] The server then uses generative AI to generate search intent for each search query stored in the database. For example, for the query "nearby shelter," the server generates intent such as "I'm looking for the specific location and access method of the shelter closest to my current location." This generated search intent is stored in the database.

[0531] Emotion Recognition Phase

[0532] Furthermore, as a unique feature of the present invention, the server utilizes an emotion engine to recognize the user's emotions. Based on the search query entered by the user, emotions such as anxiety, urgency, excitement, etc. are determined. This emotion information is incorporated into the generation of search intent.

[0533] Matching Phase

[0534] The server then automatically retrieves the actual search results for each query. This is done by using search engine APIs to retrieve the current search results for the query. For example, search for "nearby shelter" and collect the content of the top search results. The server then compares the retrieved search results with the generated search intent (including sentiment information) to determine whether they match. If the search results do not match the user's search intent, the server lists the query.

[0535] List Output Phase

[0536] This list is used to notify administrators of queries that do not match. For example, queries such as "nearby evacuation shelters," "areas with water outages," and "emergency contact information" are listed, and the server provides this list to the device, prompting the person in charge or administrator to take action.

[0537] Answer module generation phase

[0538] Finally, the server generates a tailored response module based on the list, for example, content about current evacuation shelter updates or water outages, utilizing a geographic information system (GIS) and real-time data feeds. The response module is then posted on a website or app for easy access by users.

[0539] Specific examples

[0540] Here are some specific examples from areas where earthquakes have occurred.

[0541] Assumed scenario

[0542] The user searches for "nearby shelter." The server analyzes that similar queries are frequently searched by multiple users and extracts the query "nearby shelter." It then uses a generation AI to generate the search intent: "The user is looking for information on the nearest shelter that can be reached from their current location."

[0543] At the same time, the server uses its emotion engine to recognize that the user is feeling extremely anxious, and accordingly modifies the search intent to read, "The user is in a great hurry and wants to know the exact location of the nearest evacuation shelter and how to access it quickly."

[0544] The server uses a search engine API to retrieve current search results for "nearby shelters," and if the results only contain local history and general shelter information, but lack real-time shelter location information, it determines that the query is not a match. It then lists the query and notifies the administrator.

[0545] The server generates a response module that provides the latest evacuation shelter location information and posts it on the website or app. This process allows the system to provide users with the information they need in the event of a disaster. The introduction of the emotion engine makes it possible to respond in accordance with the level of urgency and the user's emotional state, further improving the user's sense of security and convenience.

[0546] The processing flow will be explained below.

[0547] Step 1:

[0548] The server analyzes search logs related to disasters, using a technique to filter queries containing specific keywords from search engine access logs.

[0549] Step 2:

[0550] The server stores the extracted search queries in a database, including the queries themselves and their frequency of searches.

[0551] Step 3:

[0552] The server uses generative AI to generate search intent for each search query stored in the database, passing the query through an AI model that extracts from it what the user is looking for.

[0553] Step 4:

[0554] The server stores the generated search intent in a database, which identifies the specific purpose or intent associated with each query.

[0555] Step 5:

[0556] The server uses an emotion engine to recognize the user's emotions, such as anxiety, urgency, or excitement, based on the search query entered by the user.

[0557] Step 6:

[0558] The server adjusts the generated search intent based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling anxious, it determines that the user is looking for quicker and more specific information.

[0559] Step 7:

[0560] The server uses the search engine API to automatically retrieve the current search results for each query, including the content of the top-ranking pages from the search engine.

[0561] Step 8:

[0562] The server analyzes the content of the search results and compares them with the search intent (including emotional information) stored in the database. This comparison determines whether the search results match the intent.

[0563] Step 9:

[0564] The server lists queries for which search results do not match the search intent. The list includes the queries that do not match, and this information is notified to the administrator.

[0565] Step 10:

[0566] Based on the list, the server generates a response module that matches the intent, including up-to-date information on evacuation shelters and water outages.

[0567] Step 11:

[0568] The server then posts the generated answer module on a website or app, allowing users to easily access it and obtain the information they need.

[0569] Specific examples

[0570] We will explain specific examples from areas where earthquakes have occurred.

[0571] Assumed scenario

[0572] Step 1:

[0573] The user searches for "shelters near me."

[0574] Step 2:

[0575] The server analyzes that similar queries are being searched by multiple users and extracts the query "nearby shelter."

[0576] Step 3:

[0577] The server uses a generation AI to generate search intent such as, "The user is looking for information about the nearest evacuation shelter that can be reached from their current location."

[0578] Step 4:

[0579] Using its emotion engine, the server recognizes that the user is extremely anxious and modifies the search intent to read, "The user is in a great hurry and wants to know the exact location of the nearest evacuation shelter and how to access it quickly."

[0580] Step 5:

[0581] The server uses a search engine API to retrieve the current search results for "shelters near me."

[0582] Step 6:

[0583] The server collects the content of the displayed search results and determines that they contain local history and general shelter information, but do not contain real-time shelter location information.

[0584] Step 7:

[0585] The server lists queries that do not match the search intent and notifies the administrator.

[0586] Step 8:

[0587] The server generates an answer module that provides up-to-date shelter location information that matches the user's search intent.

[0588] Step 9:

[0589] The server posts the generated answer module on a website or app so that users can access it.

[0590] This process allows us to provide users with the information they need in the event of a disaster. The introduction of an emotion engine makes it possible to respond in accordance with the level of urgency and the user's emotional state, further improving the user's sense of security and convenience.

[0591] Example 2

[0592] 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."

[0593] Conventional information provision systems in the event of a disaster simply analyze the search intent of search queries without considering the user's emotions, making it difficult to provide information that reflects the level of urgency or psychological needs. Furthermore, search results often do not adequately match the user's search intent, making it necessary to provide information quickly and accurately.

[0594] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for extracting search queries related to disasters, means for generating search intent for the search queries, means for recognizing a user's emotion from the search queries, means for incorporating the emotion based on the search intent, means for acquiring search results based on the search queries, means for comparing the search results with the search intent and the emotion to determine whether they match, and means for listing search queries that do not match the search intent. This makes it possible to generate search intent taking the user's emotion into consideration, and to provide quick and accurate information that corresponds to the level of urgency and psychological needs.

[0595] A "search query" is a word or phrase that a user enters into a search engine to request specific information.

[0596] "Search intent" refers to the information or goal a user is actually seeking through a search query.

[0597] "Emotion" refers to the psychological state or feeling a user has when entering a search query.

[0598] A "server" is a computer system that provides services to other computers and terminals on a network.

[0599] A "means" is a method, device, or process for achieving a specific purpose.

[0600] "Search results" are the lists of information and links that a search engine returns in response to a user's search query.

[0601] "Listing" means selecting items that meet specific conditions as a list.

[0602] An "emotion engine" is software or algorithms that analyze and recognize emotions from a user's text or voice.

[0603] A "generative AI model" is an artificial intelligence (AI) system that is trained on large amounts of data to generate text, infer intent, recognize emotions, and more.

[0604] An "answer module" is a program or component that provides specific answer information generated based on the user's search intent and emotions.

[0605] This invention is a system for quickly and accurately providing appropriate information based on user search queries during disasters. This system incorporates an emotion engine that recognizes the user's emotions, enabling it to more accurately understand the user's search intent and provide appropriate information.

[0606] Hardware and software used

[0607] This system mainly uses the following hardware and software:

[0608] Server: A computer system for processing information.

[0609] Search engine API: An interface for retrieving search results (e.g., Google Search API, Bing Search API).

[0610] Generative AI models: AI models for text generation, intent inference, and emotion recognition (e.g., GPT-3).

[0611] Emotion engine: Software for analyzing user emotions (e.g., IBM Watson Tone Analyzer).

[0612] Database: A system for storing extracted search queries and generated intents.

[0613] NLP libraries: Natural language processing libraries (e.g., NLTK, spaCy) for analyzing search logs and processing text.

[0614] Specific operation of the system

[0615] 1. Query Extraction

[0616] The server periodically collects search engine log data. For example, it downloads the log data using a Python script and uses an NLP library to extract search queries related to disasters, such as "earthquake occurrence," "evacuation location," and "water outage."

[0617] The extracted queries are stored in a database.

[0618] 2. Generating Search Intent

[0619] The server generates search intent for the saved search query using a generative AI model (e.g., GPT-3). For example, it inputs a prompt such as "Please explain the user's intent for this search query." and stores the generated intent in a database.

[0620] 3. Emotion recognition

[0621] The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to determine emotions based on the search query entered by the user. For example, if a user searches for "nearby shelter," the emotion engine can recognize anxiety and urgency and add emotional information to the search intent.

[0622] 4. Obtain and compare search results

[0623] The server uses a search engine API (e.g., Google Search API) to obtain the current search results for the query.

[0624] The retrieved search results are compared with the generated search intent and sentiment information to determine whether they match.

[0625] 5. Listing

[0626] The server lists queries that do not match the search intent and notifies the administrator of this list, for example by outputting the mismatched queries to a CSV file and sending it by email.

[0627] 6. Generating the Answer Module

[0628] The server generates an answer module that reflects the search intent and sentiment based on the listed queries. It inputs a prompt message to the generative AI model saying, "Please generate a module that provides guidance based on the latest evacuation shelter information," and displays the generated module on a website or app.

[0629] Example scenario

[0630] For example, in an area where an earthquake has occurred, a user searches for "nearby shelter." The server analyzes search logs and extracts the query "nearby shelter." Using a generative AI model, it generates the search intent: "The user is looking for information on the nearest shelter from their current location." At the same time, it uses an emotion engine to recognize the user's anxiety and revise the search intent to "The user is in a hurry and wants to know the specific location of the nearest shelter and how to quickly access it." The server uses a search engine API to obtain current search results for "nearby shelter." If the results are inappropriate, it lists the query and notifies the administrator. Finally, the server generates an answer module that provides the latest shelter location information and posts it on a website or app, allowing users to quickly access it.

[0631] Prompt Sentence Examples

[0632] "Please explain the user's intent for this search query."

[0633] "Determine the sentiment based on the user's input query."

[0634] "Please generate a module that provides guidance based on the latest evacuation shelter information."

[0635] This system can quickly and accurately provide users with the information they need in the event of a disaster. The introduction of an emotion engine makes it possible to respond in accordance with the level of urgency and the user's emotional state, further improving user comfort and convenience.

[0636] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0637] Step 1: Query extraction phase

[0638] Input: Search engine log data

[0639] Processing: The server analyzes the search engine log data and extracts queries related to disasters. Specifically, it downloads the log data using a Python script and uses an NLP library (e.g., NLTK, spaCy) to filter queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage."

[0640] Output: A list of disaster-related search queries

[0641] Specific behavior:

[0642] The server periodically runs a Python script to retrieve log data.

[0643] The server uses an NLP library to extract queries containing keywords.

[0644] The extracted queries are stored in a database.

[0645] Step 2: Search intent generation phase

[0646] Input: A list of extracted search queries

[0647] Processing: The server uses a generative AI model (e.g., GPT-3) to generate search intent for the search query. It provides the generative AI model with a prompt, such as "Please explain the user's intent for this search query."

[0648] Output: Generated search intent

[0649] Specific behavior:

[0650] The server inputs each search query into a generative AI model.

[0651] The intent obtained from the generative AI model is stored in a database.

[0652] Step 3: Emotion Recognition Phase

[0653] Input: Extracted search query

[0654] Processing: The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. Based on the search query entered by the user, it determines emotions such as anxiety or urgency.

[0655] Output: Emotional information (score)

[0656] Specific behavior:

[0657] The server sends the search query to the emotion engine API.

[0658] The idea is to incorporate sentiment scores obtained from the sentiment engine into search intent.

[0659] Step 4: Matching Phase

[0660] Input: Search intent and sentiment information

[0661] Processing: The server uses a search engine API (e.g., Google Search API) to retrieve search results and compare them with the generated search intent and sentiment information to determine if there is a match.

[0662] Output: A list of unmatched search queries

[0663] Specific behavior:

[0664] The server sends the search query to the search engine API and retrieves the results.

[0665] The search results are analyzed and compared with search intent and sentiment information.

[0666] As a result of the comparison, queries that are determined not to match are listed.

[0667] Step 5: List output phase

[0668] Input: A list of unmatched search queries

[0669] Action: The server notifies the administrator of unmatched queries, exports the list as a CSV file, emails it, or displays it in the web dashboard.

[0670] Output: Notification to administrator (CSV file or dashboard)

[0671] Specific behavior:

[0672] The server exports unmatched queries to a CSV file.

[0673] You can either email the administrator or view the list in the web dashboard.

[0674] Step 6: Answer module generation phase

[0675] Input: Unmatched search query and search intent and sentiment information

[0676] Processing: The server generates an answer module based on the unmatched query. For example, it gives a prompt such as "Please generate a module that provides guidance based on the latest evacuation shelter information" to the generative AI model to create appropriate content.

[0677] Output: Answer module

[0678] Specific behavior:

[0679] The server inputs a prompt sentence into the generative AI model and generates an answer module.

[0680] Deploy the generated answer module to your website or app.

[0681] In this way, information based on a user's search query during a disaster can be provided quickly and accurately according to the level of urgency and the user's emotional state.

[0682] (Application example 2)

[0683] 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."

[0684] In the event of a disaster, logistics center employees need to be able to obtain information quickly and appropriately so they can continue their work safely. However, existing systems lack the ability to provide real-time inventory management and delivery information that is appropriate for the unique circumstances of a disaster, and there are also insufficient means to alleviate employee anxiety. This creates problems that make it difficult to operate efficiently.

[0685] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0686] In this invention, the server includes means for extracting search queries relevant to disasters, means for generating search intent for the search queries, means for obtaining search results based on the search queries, means for comparing the search results with the search intent to determine whether they match, means for listing search queries that do not match the search intent, means for recognizing a user's emotion regarding the search queries, means for modifying the search intent based on the emotion, means for providing inventory management and delivery information in the event of a disaster, and means for providing information access using a terminal. This enables employees at a logistics center to obtain necessary information in real time and continue their work efficiently and safely even in the event of a disaster.

[0687] A "search query" is a keyword or phrase that a user enters when searching for information.

[0688] "Search intent" is the specific information or goal a user is seeking when they enter a search query.

[0689] A "search result" is a list of relevant information returned by a search engine in response to a search query.

[0690] "Emotion" refers to the psychological state or feeling a user has when entering a search query.

[0691] "Inventory management" is the activity of understanding the quantity and condition of goods held by logistics centers and companies and managing them appropriately.

[0692] "Shipping information" refers to detailed data and information regarding the shipping of products.

[0693] A "terminal" is a device that is connected to a computer system and inputs and outputs information, such as a smartphone or a head-mounted display.

[0694] MODE FOR CARRYING OUT THE INVENTION

[0695] The present invention is a system that enables employees at a logistics center to quickly and appropriately obtain information and continue their work safely in the event of a disaster, and is composed of the following main components:

[0696] Query Extraction Phase

[0697] First, the server has a means for extracting search queries relevant to disasters. This means analyzes search queries entered by logistics center employees using smartphones or head-mounted displays, and extracts queries that are particularly important in the event of a disaster, such as "delivery interruption," "stock shortage," and "emergency route." These queries are stored in a database.

[0698] Intention generation phase

[0699] The server then uses the generative AI model to generate search intent for each search query stored in the database. For example, for the query "delivery interruption," the server generates the search intent "I'm looking for safe delivery routes and relay points in the event of a disaster." This generated search intent is stored in the database.

[0700] Emotion Recognition Phase

[0701] Furthermore, the server has a means of recognizing employees' emotions using an emotion engine. Based on the search query entered by the user, emotions such as anxiety or urgency are determined and incorporated into the generation of search intent. This allows the search intent to be modified to be more specific and in line with the urgency, for example, "The user wants to know the evacuation route quickly."

[0702] Matching Phase

[0703] Next, the server retrieves the actual search results for each query using the search engine API. It compares the retrieved search results with the generated search intent (including sentiment information) to determine whether they match. For example, if only old route information is displayed for the query "emergency route," it is determined that the query does not match and the query is listed.

[0704] List Output Phase

[0705] The server notifies the administrator of the list of unmatched queries, allowing the administrator to take prompt action, such as providing updated delivery route information or updating inventory status.

[0706] Answer module generation phase

[0707] Finally, the server generates a response module based on the list, providing up-to-date evacuation route information and inventory status, utilizing GIS and real-time data feeds. The response module is displayed on the smartphones and head-mounted displays of logistics center employees for immediate access.

[0708] Specific examples

[0709] Usage Scenarios

[0710] For example, if a logistics center employee types the query "delivery interruption" into their smartphone, the server analyzes the query and generates the search intent: "I'm looking for safe delivery routes and relay points in the event of a disaster." At the same time, the server uses its emotion engine to recognize that the employee is feeling extremely anxious and modifies the search intent to "I want to know about evacuation routes quickly." If the results obtained by the server using the search engine API contain only outdated information, it determines that there is no match and notifies the administrator. An answer module providing the latest delivery route information is then generated and displayed on the employee's device. This process allows logistics center employees to quickly obtain the information they need, even in the event of a disaster, and continue their work efficiently and safely.

[0711] Example of input prompt for generative AI model

[0712] For the query "delivery disruption," generate search intent that reflects a distribution center employee wanting the best delivery route during a disaster.

[0713] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0714] Step 1:

[0715] The server extracts disaster-related search queries by analyzing search logs and filtering queries containing specific keywords (e.g., "delivery interruption," "stock shortage," "emergency route," etc.). The input is the search logs, and the output is a list of disaster-related search queries.

[0716] Step 2:

[0717] The server uses a generative AI model to generate search intent for each extracted search query. For example, for the query "delivery interruption," the server generates the search intent "I'm looking for safe delivery routes and relay points in the event of a disaster." The input is the extracted search query, and the output is the corresponding search intent.

[0718] Step 3:

[0719] The server uses an emotion engine to recognize the emotion of the user (employee) who entered the search query. In this phase, emotions such as anxiety or urgency are determined based on the entered keywords and phrases, and information is added. The input is the search query, and the output is the search intent, including emotional characteristics.

[0720] Step 4:

[0721] The server uses the search engine API to get the current search results based on the revised search intent. Specifically, it calls the search engine API to get the "delivery interruption" information in real time. The input of this step is the revised search intent, and the output is a list of relevant search results.

[0722] Step 5:

[0723] The server compares the retrieved search results with the generated search intent and determines whether they match. For example, it checks whether the search results contain information about safe routes or intermediate stops. The input is the search results and the search intent, and the output is the result of whether they match.

[0724] Step 6:

[0725] The server lists queries that do not match the search intent and notifies the administrator, for example, when up-to-date delivery route information is missing. The input is the unmatched query, and the output is a notification message.

[0726] Step 7:

[0727] The server generates response modules based on the list, providing up-to-date evacuation route information and inventory availability, leveraging a geographic information system (GIS) and real-time data feeds. The input is the latest information or data feed, and the output is the response module.

[0728] Step 8:

[0729] The generated response module is displayed on the user's smartphone or head-mounted display, allowing employees to quickly obtain the information they need in the event of a disaster. The input is the response module, and the output is the information displayed on the device.

[0730] This will enable logistics center employees to continue working efficiently and safely even in the event of a disaster.

[0731] 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.

[0732] 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.

[0733] 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.

[0734] [Third embodiment]

[0735] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0736] 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.

[0737] 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).

[0738] 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.

[0739] 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.

[0740] 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).

[0741] 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.

[0742] 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.

[0743] 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.

[0744] 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.

[0745] 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.

[0746] 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."

[0747] The present invention provides a system for quickly and accurately providing appropriate information based on a user's search query during a disaster. The system includes the following main components:

[0748] Query Extraction Phase

[0749] First, the server has the function of extracting search queries related to disasters. This is achieved by analyzing search engine search logs and filtering relevant queries. For example, queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage" are extracted and stored in a database.

[0750] Intention generation phase

[0751] Next, the server uses generation AI to generate search intent for the extracted search query. For example, for the query "nearby shelter," it generates intent such as "I'm looking for the specific location and access method of the shelter closest to my current location." This generated search intent is stored in a database.

[0752] Matching Phase

[0753] The server then automatically retrieves the actual search results for each query. This is done by using search engine APIs to retrieve the current search results for the query. For example, search for "nearby shelter" and collect the contents of the top search results. The server then compares the retrieved search results with the generated search intent to determine whether they match. If the search results do not match the user's search intent, the server lists the query.

[0754] List Output Phase

[0755] This list is used to notify administrators of queries that do not match. For example, queries such as "nearby evacuation shelters," "areas with water outages," and "emergency contact information" are listed, and the server provides this list to the device, prompting the person in charge or administrator to take action.

[0756] Answer module generation phase

[0757] Finally, the server generates a tailored response module based on the list, for example, content about current evacuation shelter updates or water outages, utilizing a geographic information system (GIS) and real-time data feeds. The response module is then posted on a website or app for easy access by users.

[0758] Specific examples

[0759] Here are some specific examples from areas where earthquakes have occurred.

[0760] Assumed scenario

[0761] The user searches for "nearby shelter." The server analyzes that similar queries are frequently searched by multiple users and extracts the query "nearby shelter." It then uses a generation AI to generate the search intent: "The user is looking for information on the nearest shelter that can be reached from their current location."

[0762] The server uses a search engine API to retrieve current search results for "nearby shelters," and if the results only contain local history and general shelter information, but lack real-time shelter location information, it determines that the query is not a match. It then lists the query and notifies the administrator.

[0763] The server generates a response module that provides the latest evacuation shelter location information and posts it on the website or app. This process allows the system to provide users with the information they need in the event of a disaster.

[0764] The processing flow will be explained below.

[0765] Step 1:

[0766] The server analyzes search logs related to disasters, using a technique to filter queries containing specific keywords from search engine access logs.

[0767] Step 2:

[0768] The server stores the extracted search queries in a database, including the queries themselves and their frequency of searches.

[0769] Step 3:

[0770] The server uses generative AI to generate search intent for each search query stored in the database, passing the query through an AI model that extracts from it what the user is looking for.

[0771] Step 4:

[0772] The server stores the generated search intent in a database, which identifies the specific purpose or intent associated with each query.

[0773] Step 5:

[0774] The server uses the search engine API to automatically retrieve the current search results for each query, including the content of the top-ranking pages from the search engine.

[0775] Step 6:

[0776] The server analyzes the content of the search results and compares them with the search intent stored in the database to determine whether the search results match the intent.

[0777] Step 7:

[0778] The server lists queries for which search results do not match the search intent. The list includes the queries that do not match, and this information is notified to the administrator.

[0779] Step 8:

[0780] Based on the list, the server generates a response module that matches the intent, including up-to-date information on evacuation shelters and water outages.

[0781] Step 9:

[0782] The server then posts the generated answer module on a website or app, allowing users to easily access it and obtain the information they need.

[0783] Example 1

[0784] 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."

[0785] During a disaster, it is important for users to be able to access the information they need quickly and accurately. However, with conventional systems, search results often do not match the user's search intent, making it difficult to quickly obtain the information they need. Furthermore, the lack of appropriate information can increase user confusion and anxiety, creating a problem. Another issue is that information is not updated in real time, even though many users are searching for similar queries.

[0786] 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.

[0787] In this invention, the server includes means for extracting search queries related to disasters, means for generating search intent for the search queries, means for analyzing whether the search queries are searched by many users, means for saving the generated search intent in a database, means for comparing search results with the generated search intent to determine whether they match, means for listing search queries that do not match the search intent, means for notifying an administrator of the list, means for generating an answer module based on the search intent, means for displaying the answer module, and means for generating the answer module using a real-time data feed or a geographic information system. This makes it possible to quickly and accurately provide information desired by users in the event of a disaster.

[0788] A "search query" is a word or phrase that a user enters into a search engine to find information.

[0789] "Search intent" refers to the information or goal a user is really looking for when they enter a search query.

[0790] "Search results" are the list of web pages or information that a search engine displays based on a user's search query.

[0791] A "database" is a system for efficiently storing, managing, and searching large amounts of data.

[0792] A "generative AI model" is an algorithm or program that uses artificial intelligence to analyze data and make predictions or generate results that suit a specific purpose.

[0793] A "real-time data feed" is a data stream that updates with ongoing information.

[0794] A "geographic information system" is a system for collecting, managing, analyzing, and displaying geographic data and spatial information.

[0795] "Notification means" refers to a method or system for notifying relevant parties of specific information.

[0796] The present invention provides a system for providing quick and accurate information based on a user's search query during a disaster. The system includes the following main components:

[0797] 1. Query extraction phase

[0798] The server extracts search queries relevant to disasters from search engine search logs. Specifically, it uses large-scale data processing tools such as Apache Hadoop and Google BigQuery to analyze the search logs and filter queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage." The extracted queries are stored in a database (for example, MySQL or PostgreSQL).

[0799] 2. Search Intent Generation Phase

[0800] Next, the server uses a generative AI model (e.g., OpenAI's GPT-4) to generate the search intent for the extracted search query. The server inputs the query into the generative AI model and generates the intent, "The user is looking for information about the nearest shelter from their current location." This generated search intent is stored in a database. An example of an input prompt sentence is, "Please generate the user's search intent based on the search query 'nearby shelters'."

[0801] 3. Matching Phase

[0802] The server uses a search engine API (e.g., Google Search API) to retrieve current search results for a query. For the query "shelters near me," the server collects the top search results and compares them with the generated search intent. If the search results do not provide real-time shelter information, the server determines that they do not match the intent. This information is stored in a database.

[0803] 4. List output phase

[0804] The server lists search queries that do not match the search intent and notifies the administrator of this list. Notifications are sent via email or notification services (e.g., Twilio or Slack API). The administrator's dashboard displays queries such as "nearby evacuation centers," "areas with water outages," and "emergency contact information."

[0805] 5. Answer module generation phase

[0806] The server generates answer modules based on the listed queries, leveraging geographic information systems (GIS) and real-time data feeds to generate content such as up-to-date evacuation shelter locations and water outage status. The answer modules are then posted on websites and apps for easy access by users.

[0807] Specific examples

[0808] Let's take the case of an earthquake as an example. When a user searches for the query "nearby shelter," the server first analyzes the search log and extracts this query. Next, it uses a generative AI model to generate the search intent: "The user is looking for information on the shelter closest to their current location." The server retrieves search results for this query using a search engine API, and if it determines that real-time shelter information is missing, it determines that the results are not a match. This information is listed and notified to the administrator. Then, it generates an answer module based on the latest shelter location information and posts it on a website or app, allowing users to quickly and accurately access the information they need. This example ensures the accuracy and speed of information during disasters.

[0809] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0810] Step 1:

[0811] Query Extraction Phase

[0812] Input: The server receives the search logs of a search engine.

[0813] How it works: The server uses Apache Hadoop and Google BigQuery to analyze search logs. Specifically, it periodically collects search logs and filters search queries that include disaster-related keywords such as "earthquake occurrence," "evacuation location," and "water outage."

[0814] Output: The filtered search query is saved in a database.

[0815] Step 2:

[0816] Search Intent Generation Phase

[0817] Input: The server reads the extracted search query from the database.

[0818] How it works: The server inputs a query to a generative AI model (e.g., OpenAI's GPT-4) using a prompt sentence. An example of a prompt sentence is "Generate the user's search intent based on the search query 'nearby shelters'." The generative AI model analyzes the search query and generates the search intent. It generates the intent, "The user is looking for information about the shelter closest to their current location."

[0819] Output: The generated search intent is stored in a database.

[0820] Step 3:

[0821] Matching Phase

[0822] Input: The server reads the generated search intent from the database.

[0823] How it works: The server uses a search engine API (e.g., Google Search API) to retrieve current search results for a search query. The content of the retrieved search results is analyzed and compared to the generated search intent. For example, if the query "near me shelters" returns search results that are mostly about local history and general shelter information, but lack real-time shelter location information, it is determined to be a poor match.

[0824] Output: The queries that were not found to be a match are listed and stored in a database.

[0825] Step 4:

[0826] List Output Phase

[0827] Input: The server reads the listed unmatched queries.

[0828] Behavior: The server notifies the administrator of this list by running a function that displays it on the administrator's dashboard, and also notifies the administrator via an email system or notification service (e.g., Twilio, Slack API).

[0829] Output: The administrator sees a list of unmatched queries such as "nearest shelter," "area without water," and "emergency contact information."

[0830] Step 5:

[0831] Answer module generation phase

[0832] Input: The server gathers the necessary information based on the listed queries.

[0833] How it works: The server uses real-time data feeds and geographic information systems (GIS) to obtain the latest information. For example, it collects data on the latest locations of evacuation centers and water outages. It then generates answer modules based on this data. It then uses generative AI models to automatically generate content to provide users with the information they are looking for.

[0834] Output: The generated answer module is posted on a website or app for easy access by users.

[0835] (Application example 1)

[0836] 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."

[0837] In the event of a disaster, rapid and accurate information provision is required, but conventional systems sometimes have difficulty providing information that matches the user's search intent. Furthermore, in business environments such as factories, there is a lack of means for personnel to instantly obtain and display the emergency information they need. This can lead to delayed responses in the event of a disaster, potentially hindering safety.

[0838] 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.

[0839] In this invention, the server includes means for extracting search queries relevant to disasters, means for generating search intent for the search queries, means for acquiring search results based on the search queries, means for comparing the search results with the search intent to determine whether they match, means for listing search queries that do not match the search intent, means for acquiring and displaying real-time data based on the listed search queries, and means for displaying the real-time data on a user terminal. This allows for the rapid and accurate provision of information required by users, enabling rapid emergency response, particularly in business environments such as factories.

[0840] "Disaster" refers to the environment or situation when an emergency occurs due to a natural or man-made disaster.

[0841] A "search query" refers to an inquiry or keyword that a user enters into a search system.

[0842] "Search intent" refers to the purpose or true intention of a user when entering a search query.

[0843] "Search results" refers to the set of information and links that a search system returns in response to a search query.

[0844] "Match" refers to a situation where the search intent matches the search results and meets the user's needs.

[0845] "Listing" refers to extracting elements based on specific conditions and compiling them into a list.

[0846] "Real-time data" refers to the latest information at the current time or data that is updated in real time.

[0847] "User terminal" refers to a device used by a user to access information, such as a smartphone, smart glasses, or head-mounted display.

[0848] An "answer module" refers to specific information delivery methods or content generated based on search intent.

[0849] "Posting" refers to making information publicly available on a display device or platform.

[0850] The present invention is a specific embodiment of a system for providing rapid and accurate information based on user search queries during disasters. This system consists of a server, a user terminal (such as smart glasses), and a generative AI model.

[0851] Server Processing

[0852] The server extracts search queries related to disasters from search logs and generates search intent based on them. It then obtains search results based on the search queries using a search engine API and compares the obtained search results with the generated search intent. Based on this comparison, it lists search queries that do not match the user's search intent.

[0853] The server is also responsible for retrieving real-time data based on the listed search queries and displaying it on the user's device. The real-time data can come from a GIS or real-time data feeds.

[0854] User terminal processing

[0855] The user terminal, specifically the smart glasses, displays the real-time data received from the server, allowing the user to take prompt and accurate evacuation and emergency response measures in the event of a disaster.

[0856] Hardware and software used

[0857] The server is built on a cloud computing platform (e.g., AWS, Google Cloud). The search engine API can be the Google Search API. Generative AI models such as OpenAI's GPT-3 and GPT-4 are used.

[0858] The user terminal uses smart glasses (e.g., Google Glass, Vuzix Blade), which allow the user to obtain visual information in real time.

[0859] Specific examples

[0860] Consider the case where an earthquake occurs inside a factory. When a worker searches for the query "nearby evacuation site," the server analyzes the search log and extracts this query. Using a generative AI model, it generates the search intent: "The user is looking for information on the nearest evacuation site that can be reached from their current location."

[0861] Next, the server uses a search engine API to retrieve search results for "nearby evacuation sites" and determines whether the displayed results match the user's intent. If not, it lists this query and retrieves the latest evacuation site information from real-time data. This information is displayed on the smart glasses, allowing personnel to take prompt evacuation action.

[0862] Example prompts for generative AI models

[0863] Explain why users might search for "shelters near me."

[0864] In this way, it becomes possible to provide real-time information based on the user's search intent, enabling a quick and accurate response in the event of a disaster.

[0865] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0866] Step 1:

[0867] The server extracts search queries related to disasters. Specifically, the server analyzes search logs and filters search queries containing related keywords such as "earthquake," "evacuation site," and "water outage." This allows specific queries used by users during disasters to be extracted. The input is search log data, and the output is the extracted search queries.

[0868] Step 2:

[0869] The server generates search intent for the extracted search queries. It uses a generative AI model (e.g., OpenAI's GPT-3) to generate user intent for each search query. The prompt sentence is "Please explain why users search for <query>." The input is the search query, and the output is the generated search intent.

[0870] Step 3:

[0871] The server retrieves search results based on the search query. It uses a search engine API (e.g., Google Search API) to find the current search results for each search query. The input is the search query, and the output is the search results.

[0872] Step 4:

[0873] The server compares the retrieved search results with the generated search intent to determine whether they match. Specifically, it compares the content of the search results with the generated intent, and if they do not match, it lists the queries. The input is the search results and the search intent, and the output is a list of queries that do not match.

[0874] Step 5:

[0875] The server retrieves real-time data based on the listed search queries, gathers up-to-date information from GIS and real-time data feeds, and generates specific answer modules for unmatched queries. The input is a list of unmatched queries, and the output is an answer module.

[0876] Step 6:

[0877] The server displays the generated response module on the user's device. Specifically, the generated real-time data is sent to and displayed on the user's device, such as smart glasses. This allows the user to quickly obtain the necessary disaster information. The response module is the input, and the real-time data displayed on the user's device is the output.

[0878] In this way, the server performs a series of processes, making it possible to quickly and accurately provide necessary information to user terminals in the event of a disaster.

[0879] 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.

[0880] This invention is a system for quickly and accurately providing appropriate information based on user search queries during disasters. This system incorporates an emotion engine that recognizes the user's emotions, making it possible to more accurately understand the user's search intent and provide appropriate information.

[0881] Components

[0882] The system of the present invention comprises the following main components:

[0883] Query Extraction Phase

[0884] First, the server has the function of extracting search queries related to disasters. This is achieved by analyzing search engine search logs and filtering relevant queries. For example, queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage" are extracted and stored in a database.

[0885] Intention generation phase

[0886] The server then uses generative AI to generate search intent for each search query stored in the database. For example, for the query "nearby shelter," the server generates intent such as "I'm looking for the specific location and access method of the shelter closest to my current location." This generated search intent is stored in the database.

[0887] Emotion Recognition Phase

[0888] Furthermore, as a unique feature of the present invention, the server utilizes an emotion engine to recognize the user's emotions. Based on the search query entered by the user, emotions such as anxiety, urgency, excitement, etc. are determined. This emotion information is incorporated into the generation of search intent.

[0889] Matching Phase

[0890] The server then automatically retrieves the actual search results for each query. This is done by using search engine APIs to retrieve the current search results for the query. For example, search for "nearby shelter" and collect the content of the top search results. The server then compares the retrieved search results with the generated search intent (including sentiment information) to determine whether they match. If the search results do not match the user's search intent, the server lists the query.

[0891] List Output Phase

[0892] This list is used to notify administrators of queries that do not match. For example, queries such as "nearby evacuation shelters," "areas with water outages," and "emergency contact information" are listed, and the server provides this list to the device, prompting the person in charge or administrator to take action.

[0893] Answer module generation phase

[0894] Finally, the server generates a tailored response module based on the list, for example, content about current evacuation shelter updates or water outages, utilizing a geographic information system (GIS) and real-time data feeds. The response module is then posted on a website or app for easy access by users.

[0895] Specific examples

[0896] Here are some specific examples from areas where earthquakes have occurred.

[0897] Assumed scenario

[0898] The user searches for "nearby shelter." The server analyzes that similar queries are frequently searched by multiple users and extracts the query "nearby shelter." It then uses a generation AI to generate the search intent: "The user is looking for information on the nearest shelter that can be reached from their current location."

[0899] At the same time, the server uses its emotion engine to recognize that the user is feeling extremely anxious, and accordingly modifies the search intent to read, "The user is in a great hurry and wants to know the exact location of the nearest evacuation shelter and how to access it quickly."

[0900] The server uses a search engine API to retrieve current search results for "nearby shelters," and if the results only contain local history and general shelter information, but lack real-time shelter location information, it determines that the query is not a match. It then lists the query and notifies the administrator.

[0901] The server generates a response module that provides the latest evacuation shelter location information and posts it on the website or app. This process allows the system to provide users with the information they need in the event of a disaster. The introduction of the emotion engine makes it possible to respond in accordance with the level of urgency and the user's emotional state, further improving the user's sense of security and convenience.

[0902] The processing flow will be explained below.

[0903] Step 1:

[0904] The server analyzes search logs related to disasters, using a technique to filter queries containing specific keywords from search engine access logs.

[0905] Step 2:

[0906] The server stores the extracted search queries in a database, including the queries themselves and their frequency of searches.

[0907] Step 3:

[0908] The server uses generative AI to generate search intent for each search query stored in the database, passing the query through an AI model that extracts from it what the user is looking for.

[0909] Step 4:

[0910] The server stores the generated search intent in a database, which identifies the specific purpose or intent associated with each query.

[0911] Step 5:

[0912] The server uses an emotion engine to recognize the user's emotions, such as anxiety, urgency, or excitement, based on the search query entered by the user.

[0913] Step 6:

[0914] The server adjusts the generated search intent based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling anxious, it determines that the user is looking for quicker and more specific information.

[0915] Step 7:

[0916] The server uses the search engine API to automatically retrieve the current search results for each query, including the content of the top-ranking pages from the search engine.

[0917] Step 8:

[0918] The server analyzes the content of the search results and compares them with the search intent (including emotional information) stored in the database. This comparison determines whether the search results match the intent.

[0919] Step 9:

[0920] The server lists queries for which search results do not match the search intent. The list includes the queries that do not match, and this information is notified to the administrator.

[0921] Step 10:

[0922] Based on the list, the server generates a response module that matches the intent, including up-to-date information on evacuation shelters and water outages.

[0923] Step 11:

[0924] The server then posts the generated answer module on a website or app, allowing users to easily access it and obtain the information they need.

[0925] Specific examples

[0926] We will explain specific examples from areas where earthquakes have occurred.

[0927] Assumed scenario

[0928] Step 1:

[0929] The user searches for "shelters near me."

[0930] Step 2:

[0931] The server analyzes that similar queries are being searched by multiple users and extracts the query "nearby shelter."

[0932] Step 3:

[0933] The server uses a generation AI to generate search intent such as, "The user is looking for information about the nearest evacuation shelter that can be reached from their current location."

[0934] Step 4:

[0935] Using its emotion engine, the server recognizes that the user is extremely anxious and modifies the search intent to read, "The user is in a great hurry and wants to know the exact location of the nearest evacuation shelter and how to access it quickly."

[0936] Step 5:

[0937] The server uses a search engine API to retrieve the current search results for "shelters near me."

[0938] Step 6:

[0939] The server collects the content of the displayed search results and determines that they contain local history and general shelter information, but do not contain real-time shelter location information.

[0940] Step 7:

[0941] The server lists queries that do not match the search intent and notifies the administrator.

[0942] Step 8:

[0943] The server generates an answer module that provides up-to-date shelter location information that matches the user's search intent.

[0944] Step 9:

[0945] The server posts the generated answer module on a website or app so that users can access it.

[0946] This process allows us to provide users with the information they need in the event of a disaster. The introduction of an emotion engine makes it possible to respond in accordance with the level of urgency and the user's emotional state, further improving the user's sense of security and convenience.

[0947] Example 2

[0948] 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."

[0949] Conventional information provision systems in the event of a disaster simply analyze the search intent of search queries without considering the user's emotions, making it difficult to provide information that reflects the level of urgency or psychological needs. Furthermore, search results often do not adequately match the user's search intent, making it necessary to provide information quickly and accurately.

[0950] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for extracting search queries related to disasters, means for generating search intent for the search queries, means for recognizing a user's emotion from the search queries, means for incorporating the emotion based on the search intent, means for acquiring search results based on the search queries, means for comparing the search results with the search intent and the emotion to determine whether they match, and means for listing search queries that do not match the search intent. This makes it possible to generate search intent taking the user's emotion into consideration, and to provide quick and accurate information that corresponds to the level of urgency and psychological needs.

[0951] A "search query" is a word or phrase that a user enters into a search engine to request specific information.

[0952] "Search intent" refers to the information or goal a user is actually seeking through a search query.

[0953] "Emotion" refers to the psychological state or feeling a user has when entering a search query.

[0954] A "server" is a computer system that provides services to other computers and terminals on a network.

[0955] A "means" is a method, device, or process for achieving a specific purpose.

[0956] "Search results" are the lists of information and links that a search engine returns in response to a user's search query.

[0957] "Listing" means selecting items that meet specific conditions as a list.

[0958] An "emotion engine" is software or algorithms that analyze and recognize emotions from a user's text or voice.

[0959] A "generative AI model" is an artificial intelligence (AI) system that is trained on large amounts of data to generate text, infer intent, recognize emotions, and more.

[0960] An "answer module" is a program or component that provides specific answer information generated based on the user's search intent and emotions.

[0961] This invention is a system for quickly and accurately providing appropriate information based on user search queries during disasters. This system incorporates an emotion engine that recognizes the user's emotions, enabling it to more accurately understand the user's search intent and provide appropriate information.

[0962] Hardware and software used

[0963] This system mainly uses the following hardware and software:

[0964] Server: A computer system for processing information.

[0965] Search engine API: An interface for retrieving search results (e.g., Google Search API, Bing Search API).

[0966] Generative AI models: AI models for text generation, intent inference, and emotion recognition (e.g., GPT-3).

[0967] Emotion engine: Software for analyzing user emotions (e.g., IBM Watson Tone Analyzer).

[0968] Database: A system for storing extracted search queries and generated intents.

[0969] NLP libraries: Natural language processing libraries (e.g., NLTK, spaCy) for analyzing search logs and processing text.

[0970] Specific operation of the system

[0971] 1. Query Extraction

[0972] The server periodically collects search engine log data. For example, it downloads the log data using a Python script and uses an NLP library to extract search queries related to disasters, such as "earthquake occurrence," "evacuation location," and "water outage."

[0973] The extracted queries are stored in a database.

[0974] 2. Generating Search Intent

[0975] The server generates search intent for the saved search query using a generative AI model (e.g., GPT-3). For example, it inputs a prompt such as "Please explain the user's intent for this search query." and stores the generated intent in a database.

[0976] 3. Emotion recognition

[0977] The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to determine emotions based on the search query entered by the user. For example, if a user searches for "nearby shelter," the emotion engine can recognize anxiety and urgency and add emotional information to the search intent.

[0978] 4. Obtain and compare search results

[0979] The server uses a search engine API (e.g., Google Search API) to obtain the current search results for the query.

[0980] The retrieved search results are compared with the generated search intent and sentiment information to determine whether they match.

[0981] 5. Listing

[0982] The server lists queries that do not match the search intent and notifies the administrator of this list, for example by outputting the mismatched queries to a CSV file and sending it by email.

[0983] 6. Generating the Answer Module

[0984] The server generates an answer module that reflects the search intent and sentiment based on the listed queries. It inputs a prompt message to the generative AI model saying, "Please generate a module that provides guidance based on the latest evacuation shelter information," and displays the generated module on a website or app.

[0985] Example scenario

[0986] For example, in an area where an earthquake has occurred, a user searches for "nearby shelter." The server analyzes search logs and extracts the query "nearby shelter." Using a generative AI model, it generates the search intent: "The user is looking for information on the nearest shelter from their current location." At the same time, it uses an emotion engine to recognize the user's anxiety and revise the search intent to "The user is in a hurry and wants to know the specific location of the nearest shelter and how to quickly access it." The server uses a search engine API to obtain current search results for "nearby shelter." If the results are inappropriate, it lists the query and notifies the administrator. Finally, the server generates an answer module that provides the latest shelter location information and posts it on a website or app, allowing users to quickly access it.

[0987] Prompt Sentence Examples

[0988] "Please explain the user's intent for this search query."

[0989] "Determine the sentiment based on the user's input query."

[0990] "Please generate a module that provides guidance based on the latest evacuation shelter information."

[0991] This system can quickly and accurately provide users with the information they need in the event of a disaster. The introduction of an emotion engine makes it possible to respond in accordance with the level of urgency and the user's emotional state, further improving user comfort and convenience.

[0992] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0993] Step 1: Query extraction phase

[0994] Input: Search engine log data

[0995] Processing: The server analyzes the search engine log data and extracts queries related to disasters. Specifically, it downloads the log data using a Python script and uses an NLP library (e.g., NLTK, spaCy) to filter queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage."

[0996] Output: A list of disaster-related search queries

[0997] Specific behavior:

[0998] The server periodically runs a Python script to retrieve log data.

[0999] The server uses an NLP library to extract queries containing keywords.

[1000] The extracted queries are stored in a database.

[1001] Step 2: Search intent generation phase

[1002] Input: A list of extracted search queries

[1003] Processing: The server uses a generative AI model (e.g., GPT-3) to generate search intent for the search query. It provides the generative AI model with a prompt, such as "Please explain the user's intent for this search query."

[1004] Output: Generated search intent

[1005] Specific behavior:

[1006] The server inputs each search query into a generative AI model.

[1007] The intent obtained from the generative AI model is stored in a database.

[1008] Step 3: Emotion Recognition Phase

[1009] Input: Extracted search query

[1010] Processing: The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. Based on the search query entered by the user, it determines emotions such as anxiety or urgency.

[1011] Output: Emotional information (score)

[1012] Specific behavior:

[1013] The server sends the search query to the emotion engine API.

[1014] The idea is to incorporate sentiment scores obtained from the sentiment engine into search intent.

[1015] Step 4: Matching Phase

[1016] Input: Search intent and sentiment information

[1017] Processing: The server uses a search engine API (e.g., Google Search API) to retrieve search results and compare them with the generated search intent and sentiment information to determine if there is a match.

[1018] Output: A list of unmatched search queries

[1019] Specific behavior:

[1020] The server sends the search query to the search engine API and retrieves the results.

[1021] The search results are analyzed and compared with search intent and sentiment information.

[1022] As a result of the comparison, queries that are determined not to match are listed.

[1023] Step 5: List output phase

[1024] Input: A list of unmatched search queries

[1025] Action: The server notifies the administrator of unmatched queries, exports the list as a CSV file, emails it, or displays it in the web dashboard.

[1026] Output: Notification to administrator (CSV file or dashboard)

[1027] Specific behavior:

[1028] The server exports unmatched queries to a CSV file.

[1029] You can either email the administrator or view the list in the web dashboard.

[1030] Step 6: Answer module generation phase

[1031] Input: Unmatched search query and search intent and sentiment information

[1032] Processing: The server generates an answer module based on the unmatched query. For example, it gives a prompt such as "Please generate a module that provides guidance based on the latest evacuation shelter information" to the generative AI model to create appropriate content.

[1033] Output: Answer module

[1034] Specific behavior:

[1035] The server inputs a prompt sentence into the generative AI model and generates an answer module.

[1036] Deploy the generated answer module to your website or app.

[1037] In this way, information based on a user's search query during a disaster can be provided quickly and accurately according to the level of urgency and the user's emotional state.

[1038] (Application example 2)

[1039] 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."

[1040] In the event of a disaster, logistics center employees need to be able to obtain information quickly and appropriately so they can continue their work safely. However, existing systems lack the ability to provide real-time inventory management and delivery information that is appropriate for the unique circumstances of a disaster, and there are also insufficient means to alleviate employee anxiety. This creates problems that make it difficult to operate efficiently.

[1041] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1042] In this invention, the server includes means for extracting search queries relevant to disasters, means for generating search intent for the search queries, means for obtaining search results based on the search queries, means for comparing the search results with the search intent to determine whether they match, means for listing search queries that do not match the search intent, means for recognizing a user's emotion regarding the search queries, means for modifying the search intent based on the emotion, means for providing inventory management and delivery information in the event of a disaster, and means for providing information access using a terminal. This enables employees at a logistics center to obtain necessary information in real time and continue their work efficiently and safely even in the event of a disaster.

[1043] A "search query" is a keyword or phrase that a user enters when searching for information.

[1044] "Search intent" is the specific information or goal a user is seeking when they enter a search query.

[1045] A "search result" is a list of relevant information returned by a search engine in response to a search query.

[1046] "Emotion" refers to the psychological state or feeling a user has when entering a search query.

[1047] "Inventory management" is the activity of understanding the quantity and condition of goods held by logistics centers and companies and managing them appropriately.

[1048] "Shipping information" refers to detailed data and information regarding the shipping of products.

[1049] A "terminal" is a device that is connected to a computer system and inputs and outputs information, such as a smartphone or a head-mounted display.

[1050] MODE FOR CARRYING OUT THE INVENTION

[1051] The present invention is a system that enables employees at a logistics center to quickly and appropriately obtain information and continue their work safely in the event of a disaster, and is composed of the following main components:

[1052] Query Extraction Phase

[1053] First, the server has a means for extracting search queries relevant to disasters. This means analyzes search queries entered by logistics center employees using smartphones or head-mounted displays, and extracts queries that are particularly important in the event of a disaster, such as "delivery interruption," "stock shortage," and "emergency route." These queries are stored in a database.

[1054] Intention generation phase

[1055] The server then uses the generative AI model to generate search intent for each search query stored in the database. For example, for the query "delivery interruption," the server generates the search intent "I'm looking for safe delivery routes and relay points in the event of a disaster." This generated search intent is stored in the database.

[1056] Emotion Recognition Phase

[1057] Furthermore, the server has a means of recognizing employees' emotions using an emotion engine. Based on the search query entered by the user, emotions such as anxiety or urgency are determined and incorporated into the generation of search intent. This allows the search intent to be modified to be more specific and in line with the urgency, for example, "The user wants to know the evacuation route quickly."

[1058] Matching Phase

[1059] Next, the server retrieves the actual search results for each query using the search engine API. It compares the retrieved search results with the generated search intent (including sentiment information) to determine whether they match. For example, if only old route information is displayed for the query "emergency route," it is determined that the query does not match and the query is listed.

[1060] List Output Phase

[1061] The server notifies the administrator of the list of unmatched queries, allowing the administrator to take prompt action, such as providing updated delivery route information or updating inventory status.

[1062] Answer module generation phase

[1063] Finally, the server generates a response module based on the list, providing up-to-date evacuation route information and inventory status, utilizing GIS and real-time data feeds. The response module is displayed on the smartphones and head-mounted displays of logistics center employees for immediate access.

[1064] Specific examples

[1065] Usage Scenarios

[1066] For example, if a logistics center employee types the query "delivery interruption" into their smartphone, the server analyzes the query and generates the search intent: "I'm looking for safe delivery routes and relay points in the event of a disaster." At the same time, the server uses its emotion engine to recognize that the employee is feeling extremely anxious and modifies the search intent to "I want to know about evacuation routes quickly." If the results obtained by the server using the search engine API contain only outdated information, it determines that there is no match and notifies the administrator. An answer module providing the latest delivery route information is then generated and displayed on the employee's device. This process allows logistics center employees to quickly obtain the information they need, even in the event of a disaster, and continue their work efficiently and safely.

[1067] Example of input prompt for generative AI model

[1068] For the query "delivery disruption," generate search intent that reflects a distribution center employee wanting the best delivery route during a disaster.

[1069] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1070] Step 1:

[1071] The server extracts disaster-related search queries by analyzing search logs and filtering queries containing specific keywords (e.g., "delivery interruption," "stock shortage," "emergency route," etc.). The input is the search logs, and the output is a list of disaster-related search queries.

[1072] Step 2:

[1073] The server uses a generative AI model to generate search intent for each extracted search query. For example, for the query "delivery interruption," the server generates the search intent "I'm looking for safe delivery routes and relay points in the event of a disaster." The input is the extracted search query, and the output is the corresponding search intent.

[1074] Step 3:

[1075] The server uses an emotion engine to recognize the emotion of the user (employee) who entered the search query. In this phase, emotions such as anxiety or urgency are determined based on the entered keywords and phrases, and information is added. The input is the search query, and the output is the search intent, including emotional characteristics.

[1076] Step 4:

[1077] The server uses the search engine API to get the current search results based on the revised search intent. Specifically, it calls the search engine API to get the "delivery interruption" information in real time. The input of this step is the revised search intent, and the output is a list of relevant search results.

[1078] Step 5:

[1079] The server compares the retrieved search results with the generated search intent and determines whether they match. For example, it checks whether the search results contain information about safe routes or intermediate stops. The input is the search results and the search intent, and the output is the result of whether they match.

[1080] Step 6:

[1081] The server lists queries that do not match the search intent and notifies the administrator, for example, when up-to-date delivery route information is missing. The input is the unmatched query, and the output is a notification message.

[1082] Step 7:

[1083] The server generates response modules based on the list, providing up-to-date evacuation route information and inventory availability, leveraging a geographic information system (GIS) and real-time data feeds. The input is the latest information or data feed, and the output is the response module.

[1084] Step 8:

[1085] The generated response module is displayed on the user's smartphone or head-mounted display, allowing employees to quickly obtain the information they need in the event of a disaster. The input is the response module, and the output is the information displayed on the device.

[1086] This will enable logistics center employees to continue working efficiently and safely even in the event of a disaster.

[1087] 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.

[1088] 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.

[1089] 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.

[1090] [Fourth embodiment]

[1091] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1092] 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.

[1093] 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).

[1094] 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.

[1095] 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.

[1096] 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).

[1097] 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.

[1098] 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.

[1099] 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.

[1100] 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.

[1101] 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.

[1102] 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.

[1103] 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."

[1104] The present invention provides a system for quickly and accurately providing appropriate information based on a user's search query during a disaster. The system includes the following main components:

[1105] Query Extraction Phase

[1106] First, the server has the function of extracting search queries related to disasters. This is achieved by analyzing search engine search logs and filtering relevant queries. For example, queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage" are extracted and stored in a database.

[1107] Intention generation phase

[1108] Next, the server uses generation AI to generate search intent for the extracted search query. For example, for the query "nearby shelter," it generates intent such as "I'm looking for the specific location and access method of the shelter closest to my current location." This generated search intent is stored in a database.

[1109] Matching Phase

[1110] The server then automatically retrieves the actual search results for each query. This is done by using search engine APIs to retrieve the current search results for the query. For example, search for "nearby shelter" and collect the contents of the top search results. The server then compares the retrieved search results with the generated search intent to determine whether they match. If the search results do not match the user's search intent, the server lists the query.

[1111] List Output Phase

[1112] This list is used to notify administrators of queries that do not match. For example, queries such as "nearby evacuation shelters," "areas with water outages," and "emergency contact information" are listed, and the server provides this list to the device, prompting the person in charge or administrator to take action.

[1113] Answer module generation phase

[1114] Finally, the server generates a tailored response module based on the list, for example, content about current evacuation shelter updates or water outages, utilizing a geographic information system (GIS) and real-time data feeds. The response module is then posted on a website or app for easy access by users.

[1115] Specific examples

[1116] Here are some specific examples from areas where earthquakes have occurred.

[1117] Assumed scenario

[1118] The user searches for "nearby shelter." The server analyzes that similar queries are frequently searched by multiple users and extracts the query "nearby shelter." It then uses a generation AI to generate the search intent: "The user is looking for information on the nearest shelter that can be reached from their current location."

[1119] The server uses a search engine API to retrieve current search results for "nearby shelters," and if the results only contain local history and general shelter information, but lack real-time shelter location information, it determines that the query is not a match. It then lists the query and notifies the administrator.

[1120] The server generates a response module that provides the latest evacuation shelter location information and posts it on the website or app. This process allows the system to provide users with the information they need in the event of a disaster.

[1121] The processing flow will be explained below.

[1122] Step 1:

[1123] The server analyzes search logs related to disasters, using a technique to filter queries containing specific keywords from search engine access logs.

[1124] Step 2:

[1125] The server stores the extracted search queries in a database, including the queries themselves and their frequency of searches.

[1126] Step 3:

[1127] The server uses generative AI to generate search intent for each search query stored in the database, passing the query through an AI model that extracts from it what the user is looking for.

[1128] Step 4:

[1129] The server stores the generated search intent in a database, which identifies the specific purpose or intent associated with each query.

[1130] Step 5:

[1131] The server uses the search engine API to automatically retrieve the current search results for each query, including the content of the top-ranking pages from the search engine.

[1132] Step 6:

[1133] The server analyzes the content of the search results and compares them with the search intent stored in the database to determine whether the search results match the intent.

[1134] Step 7:

[1135] The server lists queries for which search results do not match the search intent. The list includes the queries that do not match, and this information is notified to the administrator.

[1136] Step 8:

[1137] Based on the list, the server generates a response module that matches the intent, including up-to-date information on evacuation shelters and water outages.

[1138] Step 9:

[1139] The server then posts the generated answer module on a website or app, allowing users to easily access it and obtain the information they need.

[1140] Example 1

[1141] 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."

[1142] During a disaster, it is important for users to be able to access the information they need quickly and accurately. However, with conventional systems, search results often do not match the user's search intent, making it difficult to quickly obtain the information they need. Furthermore, the lack of appropriate information can increase user confusion and anxiety, creating a problem. Another issue is that information is not updated in real time, even though many users are searching for similar queries.

[1143] 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.

[1144] In this invention, the server includes means for extracting search queries related to disasters, means for generating search intent for the search queries, means for analyzing whether the search queries are searched by many users, means for saving the generated search intent in a database, means for comparing search results with the generated search intent to determine whether they match, means for listing search queries that do not match the search intent, means for notifying an administrator of the list, means for generating an answer module based on the search intent, means for displaying the answer module, and means for generating the answer module using a real-time data feed or a geographic information system. This makes it possible to quickly and accurately provide information desired by users in the event of a disaster.

[1145] A "search query" is a word or phrase that a user enters into a search engine to find information.

[1146] "Search intent" refers to the information or goal a user is really looking for when they enter a search query.

[1147] "Search results" are the list of web pages or information that a search engine displays based on a user's search query.

[1148] A "database" is a system for efficiently storing, managing, and searching large amounts of data.

[1149] A "generative AI model" is an algorithm or program that uses artificial intelligence to analyze data and make predictions or generate results that suit a specific purpose.

[1150] A "real-time data feed" is a data stream that updates with ongoing information.

[1151] A "geographic information system" is a system for collecting, managing, analyzing, and displaying geographic data and spatial information.

[1152] "Notification means" refers to a method or system for notifying relevant parties of specific information.

[1153] The present invention provides a system for providing quick and accurate information based on a user's search query during a disaster. The system includes the following main components:

[1154] 1. Query extraction phase

[1155] The server extracts search queries relevant to disasters from search engine search logs. Specifically, it uses large-scale data processing tools such as Apache Hadoop and Google BigQuery to analyze the search logs and filter queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage." The extracted queries are stored in a database (for example, MySQL or PostgreSQL).

[1156] 2. Search Intent Generation Phase

[1157] Next, the server uses a generative AI model (e.g., OpenAI's GPT-4) to generate the search intent for the extracted search query. The server inputs the query into the generative AI model and generates the intent, "The user is looking for information about the nearest shelter from their current location." This generated search intent is stored in a database. An example of an input prompt sentence is, "Please generate the user's search intent based on the search query 'nearby shelters'."

[1158] 3. Matching Phase

[1159] The server uses a search engine API (e.g., Google Search API) to retrieve current search results for a query. For the query "shelters near me," the server collects the top search results and compares them with the generated search intent. If the search results do not provide real-time shelter information, the server determines that they do not match the intent. This information is stored in a database.

[1160] 4. List output phase

[1161] The server lists search queries that do not match the search intent and notifies the administrator of this list. Notifications are sent via email or notification services (e.g., Twilio or Slack API). The administrator's dashboard displays queries such as "nearby evacuation centers," "areas with water outages," and "emergency contact information."

[1162] 5. Answer module generation phase

[1163] The server generates answer modules based on the listed queries, leveraging geographic information systems (GIS) and real-time data feeds to generate content such as up-to-date evacuation shelter locations and water outage status. The answer modules are then posted on websites and apps for easy access by users.

[1164] Specific examples

[1165] Let's take the case of an earthquake as an example. When a user searches for the query "nearby shelter," the server first analyzes the search log and extracts this query. Next, it uses a generative AI model to generate the search intent: "The user is looking for information on the shelter closest to their current location." The server retrieves search results for this query using a search engine API, and if it determines that real-time shelter information is missing, it determines that the results are not a match. This information is listed and notified to the administrator. Then, it generates an answer module based on the latest shelter location information and posts it on a website or app, allowing users to quickly and accurately access the information they need. This example ensures the accuracy and speed of information during disasters.

[1166] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1167] Step 1:

[1168] Query Extraction Phase

[1169] Input: The server receives the search logs of a search engine.

[1170] How it works: The server uses Apache Hadoop and Google BigQuery to analyze search logs. Specifically, it periodically collects search logs and filters search queries that include disaster-related keywords such as "earthquake occurrence," "evacuation location," and "water outage."

[1171] Output: The filtered search query is saved in a database.

[1172] Step 2:

[1173] Search Intent Generation Phase

[1174] Input: The server reads the extracted search query from the database.

[1175] How it works: The server inputs a query to a generative AI model (e.g., OpenAI's GPT-4) using a prompt sentence. An example of a prompt sentence is "Generate the user's search intent based on the search query 'nearby shelters'." The generative AI model analyzes the search query and generates the search intent. It generates the intent, "The user is looking for information about the shelter closest to their current location."

[1176] Output: The generated search intent is stored in a database.

[1177] Step 3:

[1178] Matching Phase

[1179] Input: The server reads the generated search intent from the database.

[1180] How it works: The server uses a search engine API (e.g., Google Search API) to retrieve current search results for a search query. The content of the retrieved search results is analyzed and compared to the generated search intent. For example, if the query "near me shelters" returns search results that are mostly about local history and general shelter information, but lack real-time shelter location information, it is determined to be a poor match.

[1181] Output: The queries that were not found to be a match are listed and stored in a database.

[1182] Step 4:

[1183] List Output Phase

[1184] Input: The server reads the listed unmatched queries.

[1185] Behavior: The server notifies the administrator of this list by running a function that displays it on the administrator's dashboard, and also notifies the administrator via an email system or notification service (e.g., Twilio, Slack API).

[1186] Output: The administrator sees a list of unmatched queries such as "nearest shelter," "area without water," and "emergency contact information."

[1187] Step 5:

[1188] Answer module generation phase

[1189] Input: The server gathers the necessary information based on the listed queries.

[1190] How it works: The server uses real-time data feeds and geographic information systems (GIS) to obtain the latest information. For example, it collects data on the latest locations of evacuation centers and water outages. It then generates answer modules based on this data. It then uses generative AI models to automatically generate content to provide users with the information they are looking for.

[1191] Output: The generated answer module is posted on a website or app for easy access by users.

[1192] (Application example 1)

[1193] 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."

[1194] In the event of a disaster, rapid and accurate information provision is required, but conventional systems sometimes have difficulty providing information that matches the user's search intent. Furthermore, in business environments such as factories, there is a lack of means for personnel to instantly obtain and display the emergency information they need. This can lead to delayed responses in the event of a disaster, potentially hindering safety.

[1195] 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.

[1196] In this invention, the server includes means for extracting search queries relevant to disasters, means for generating search intent for the search queries, means for acquiring search results based on the search queries, means for comparing the search results with the search intent to determine whether they match, means for listing search queries that do not match the search intent, means for acquiring and displaying real-time data based on the listed search queries, and means for displaying the real-time data on a user terminal. This allows for the rapid and accurate provision of information required by users, enabling rapid emergency response, particularly in business environments such as factories.

[1197] "Disaster" refers to the environment or situation when an emergency occurs due to a natural or man-made disaster.

[1198] A "search query" refers to an inquiry or keyword that a user enters into a search system.

[1199] "Search intent" refers to the purpose or true intention of a user when entering a search query.

[1200] "Search results" refers to the set of information and links that a search system returns in response to a search query.

[1201] "Match" refers to a situation where the search intent matches the search results and meets the user's needs.

[1202] "Listing" refers to extracting elements based on specific conditions and compiling them into a list.

[1203] "Real-time data" refers to the latest information at the current time or data that is updated in real time.

[1204] "User terminal" refers to a device used by a user to access information, such as a smartphone, smart glasses, or head-mounted display.

[1205] An "answer module" refers to specific information delivery methods or content generated based on search intent.

[1206] "Posting" refers to making information publicly available on a display device or platform.

[1207] The present invention is a specific embodiment of a system for providing rapid and accurate information based on user search queries during disasters. This system consists of a server, a user terminal (such as smart glasses), and a generative AI model.

[1208] Server Processing

[1209] The server extracts search queries related to disasters from search logs and generates search intent based on them. It then obtains search results based on the search queries using a search engine API and compares the obtained search results with the generated search intent. Based on this comparison, it lists search queries that do not match the user's search intent.

[1210] The server is also responsible for retrieving real-time data based on the listed search queries and displaying it on the user's device. The real-time data can come from a GIS or real-time data feeds.

[1211] User terminal processing

[1212] The user terminal, specifically the smart glasses, displays the real-time data received from the server, allowing the user to take prompt and accurate evacuation and emergency response measures in the event of a disaster.

[1213] Hardware and software used

[1214] The server is built on a cloud computing platform (e.g., AWS, Google Cloud). The search engine API can be the Google Search API. Generative AI models such as OpenAI's GPT-3 and GPT-4 are used.

[1215] The user terminal uses smart glasses (e.g., Google Glass, Vuzix Blade), which allow the user to obtain visual information in real time.

[1216] Specific examples

[1217] Consider the case where an earthquake occurs inside a factory. When a worker searches for the query "nearby evacuation site," the server analyzes the search log and extracts this query. Using a generative AI model, it generates the search intent: "The user is looking for information on the nearest evacuation site that can be reached from their current location."

[1218] Next, the server uses a search engine API to retrieve search results for "nearby evacuation sites" and determines whether the displayed results match the user's intent. If not, it lists this query and retrieves the latest evacuation site information from real-time data. This information is displayed on the smart glasses, allowing personnel to take prompt evacuation action.

[1219] Example prompts for generative AI models

[1220] Explain why users might search for "shelters near me."

[1221] In this way, it becomes possible to provide real-time information based on the user's search intent, enabling a quick and accurate response in the event of a disaster.

[1222] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1223] Step 1:

[1224] The server extracts search queries related to disasters. Specifically, the server analyzes search logs and filters search queries containing related keywords such as "earthquake," "evacuation site," and "water outage." This allows specific queries used by users during disasters to be extracted. The input is search log data, and the output is the extracted search queries.

[1225] Step 2:

[1226] The server generates search intent for the extracted search queries. It uses a generative AI model (e.g., OpenAI's GPT-3) to generate user intent for each search query. The prompt sentence is "Please explain why users search for <query>." The input is the search query, and the output is the generated search intent.

[1227] Step 3:

[1228] The server retrieves search results based on the search query. It uses a search engine API (e.g., Google Search API) to find the current search results for each search query. The input is the search query, and the output is the search results.

[1229] Step 4:

[1230] The server compares the retrieved search results with the generated search intent to determine whether they match. Specifically, it compares the content of the search results with the generated intent, and if they do not match, it lists the queries. The input is the search results and the search intent, and the output is a list of queries that do not match.

[1231] Step 5:

[1232] The server retrieves real-time data based on the listed search queries, gathers up-to-date information from GIS and real-time data feeds, and generates specific answer modules for unmatched queries. The input is a list of unmatched queries, and the output is an answer module.

[1233] Step 6:

[1234] The server displays the generated response module on the user's device. Specifically, the generated real-time data is sent to and displayed on the user's device, such as smart glasses. This allows the user to quickly obtain the necessary disaster information. The response module is the input, and the real-time data displayed on the user's device is the output.

[1235] In this way, the server performs a series of processes, making it possible to quickly and accurately provide necessary information to user terminals in the event of a disaster.

[1236] 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.

[1237] This invention is a system for quickly and accurately providing appropriate information based on user search queries during disasters. This system incorporates an emotion engine that recognizes the user's emotions, making it possible to more accurately understand the user's search intent and provide appropriate information.

[1238] Components

[1239] The system of the present invention comprises the following main components:

[1240] Query Extraction Phase

[1241] First, the server has the function of extracting search queries related to disasters. This is achieved by analyzing search engine search logs and filtering relevant queries. For example, queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage" are extracted and stored in a database.

[1242] Intention generation phase

[1243] The server then uses generative AI to generate search intent for each search query stored in the database. For example, for the query "nearby shelter," the server generates intent such as "I'm looking for the specific location and access method of the shelter closest to my current location." This generated search intent is stored in the database.

[1244] Emotion Recognition Phase

[1245] Furthermore, as a unique feature of the present invention, the server utilizes an emotion engine to recognize the user's emotions. Based on the search query entered by the user, emotions such as anxiety, urgency, excitement, etc. are determined. This emotion information is incorporated into the generation of search intent.

[1246] Matching Phase

[1247] The server then automatically retrieves the actual search results for each query. This is done by using search engine APIs to retrieve the current search results for the query. For example, search for "nearby shelter" and collect the content of the top search results. The server then compares the retrieved search results with the generated search intent (including sentiment information) to determine whether they match. If the search results do not match the user's search intent, the server lists the query.

[1248] List Output Phase

[1249] This list is used to notify administrators of queries that do not match. For example, queries such as "nearby evacuation shelters," "areas with water outages," and "emergency contact information" are listed, and the server provides this list to the device, prompting the person in charge or administrator to take action.

[1250] Answer module generation phase

[1251] Finally, the server generates a tailored response module based on the list, for example, content about current evacuation shelter updates or water outages, utilizing a geographic information system (GIS) and real-time data feeds. The response module is then posted on a website or app for easy access by users.

[1252] Specific examples

[1253] Here are some specific examples from areas where earthquakes have occurred.

[1254] Assumed scenario

[1255] The user searches for "nearby shelter." The server analyzes that similar queries are frequently searched by multiple users and extracts the query "nearby shelter." It then uses a generation AI to generate the search intent: "The user is looking for information on the nearest shelter that can be reached from their current location."

[1256] At the same time, the server uses its emotion engine to recognize that the user is feeling extremely anxious, and accordingly modifies the search intent to read, "The user is in a great hurry and wants to know the exact location of the nearest evacuation shelter and how to access it quickly."

[1257] The server uses a search engine API to retrieve current search results for "nearby shelters," and if the results only contain local history and general shelter information, but lack real-time shelter location information, it determines that the query is not a match. It then lists the query and notifies the administrator.

[1258] The server generates a response module that provides the latest evacuation shelter location information and posts it on the website or app. This process allows the system to provide users with the information they need in the event of a disaster. The introduction of the emotion engine makes it possible to respond in accordance with the level of urgency and the user's emotional state, further improving the user's sense of security and convenience.

[1259] The processing flow will be explained below.

[1260] Step 1:

[1261] The server analyzes search logs related to disasters, using a technique to filter queries containing specific keywords from search engine access logs.

[1262] Step 2:

[1263] The server stores the extracted search queries in a database, including the queries themselves and their frequency of searches.

[1264] Step 3:

[1265] The server uses generative AI to generate search intent for each search query stored in the database, passing the query through an AI model that extracts from it what the user is looking for.

[1266] Step 4:

[1267] The server stores the generated search intent in a database, which identifies the specific purpose or intent associated with each query.

[1268] Step 5:

[1269] The server uses an emotion engine to recognize the user's emotions, such as anxiety, urgency, or excitement, based on the search query entered by the user.

[1270] Step 6:

[1271] The server adjusts the generated search intent based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling anxious, it determines that the user is looking for quicker and more specific information.

[1272] Step 7:

[1273] The server uses the search engine API to automatically retrieve the current search results for each query, including the content of the top-ranking pages from the search engine.

[1274] Step 8:

[1275] The server analyzes the content of the search results and compares them with the search intent (including emotional information) stored in the database. This comparison determines whether the search results match the intent.

[1276] Step 9:

[1277] The server lists queries for which search results do not match the search intent. The list includes the queries that do not match, and this information is notified to the administrator.

[1278] Step 10:

[1279] Based on the list, the server generates a response module that matches the intent, including up-to-date information on evacuation shelters and water outages.

[1280] Step 11:

[1281] The server then posts the generated answer module on a website or app, allowing users to easily access it and obtain the information they need.

[1282] Specific examples

[1283] We will explain specific examples from areas where earthquakes have occurred.

[1284] Assumed scenario

[1285] Step 1:

[1286] The user searches for "shelters near me."

[1287] Step 2:

[1288] The server analyzes that similar queries are being searched by multiple users and extracts the query "nearby shelter."

[1289] Step 3:

[1290] The server uses a generation AI to generate search intent such as, "The user is looking for information about the nearest evacuation shelter that can be reached from their current location."

[1291] Step 4:

[1292] Using its emotion engine, the server recognizes that the user is extremely anxious and modifies the search intent to read, "The user is in a great hurry and wants to know the exact location of the nearest evacuation shelter and how to access it quickly."

[1293] Step 5:

[1294] The server uses a search engine API to retrieve the current search results for "shelters near me."

[1295] Step 6:

[1296] The server collects the content of the displayed search results and determines that they contain local history and general shelter information, but do not contain real-time shelter location information.

[1297] Step 7:

[1298] The server lists queries that do not match the search intent and notifies the administrator.

[1299] Step 8:

[1300] The server generates an answer module that provides up-to-date shelter location information that matches the user's search intent.

[1301] Step 9:

[1302] The server posts the generated answer module on a website or app so that users can access it.

[1303] This process allows us to provide users with the information they need in the event of a disaster. The introduction of an emotion engine makes it possible to respond in accordance with the level of urgency and the user's emotional state, further improving the user's sense of security and convenience.

[1304] Example 2

[1305] 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."

[1306] Conventional information provision systems in the event of a disaster simply analyze the search intent of search queries without considering the user's emotions, making it difficult to provide information that reflects the level of urgency or psychological needs. Furthermore, search results often do not adequately match the user's search intent, making it necessary to provide information quickly and accurately.

[1307] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for extracting search queries related to disasters, means for generating search intent for the search queries, means for recognizing a user's emotion from the search queries, means for incorporating the emotion based on the search intent, means for acquiring search results based on the search queries, means for comparing the search results with the search intent and the emotion to determine whether they match, and means for listing search queries that do not match the search intent. This makes it possible to generate search intent taking the user's emotion into consideration, and to provide quick and accurate information that corresponds to the level of urgency and psychological needs.

[1308] A "search query" is a word or phrase that a user enters into a search engine to request specific information.

[1309] "Search intent" refers to the information or goal a user is actually seeking through a search query.

[1310] "Emotion" refers to the psychological state or feeling a user has when entering a search query.

[1311] A "server" is a computer system that provides services to other computers and terminals on a network.

[1312] A "means" is a method, device, or process for achieving a specific purpose.

[1313] "Search results" are the lists of information and links that a search engine returns in response to a user's search query.

[1314] "Listing" means selecting items that meet specific conditions as a list.

[1315] An "emotion engine" is software or algorithms that analyze and recognize emotions from a user's text or voice.

[1316] A "generative AI model" is an artificial intelligence (AI) system that is trained on large amounts of data to generate text, infer intent, recognize emotions, and more.

[1317] An "answer module" is a program or component that provides specific answer information generated based on the user's search intent and emotions.

[1318] This invention is a system for quickly and accurately providing appropriate information based on user search queries during disasters. This system incorporates an emotion engine that recognizes the user's emotions, enabling it to more accurately understand the user's search intent and provide appropriate information.

[1319] Hardware and software used

[1320] This system mainly uses the following hardware and software:

[1321] Server: A computer system for processing information.

[1322] Search engine API: An interface for retrieving search results (e.g., Google Search API, Bing Search API).

[1323] Generative AI models: AI models for text generation, intent inference, and emotion recognition (e.g., GPT-3).

[1324] Emotion engine: Software for analyzing user emotions (e.g., IBM Watson Tone Analyzer).

[1325] Database: A system for storing extracted search queries and generated intents.

[1326] NLP libraries: Natural language processing libraries (e.g., NLTK, spaCy) for analyzing search logs and processing text.

[1327] Specific operation of the system

[1328] 1. Query Extraction

[1329] The server periodically collects search engine log data. For example, it downloads the log data using a Python script and uses an NLP library to extract search queries related to disasters, such as "earthquake occurrence," "evacuation location," and "water outage."

[1330] The extracted queries are stored in a database.

[1331] 2. Generating Search Intent

[1332] The server generates search intent for the saved search query using a generative AI model (e.g., GPT-3). For example, it inputs a prompt such as "Please explain the user's intent for this search query." and stores the generated intent in a database.

[1333] 3. Emotion recognition

[1334] The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to determine emotions based on the search query entered by the user. For example, if a user searches for "nearby shelter," the emotion engine can recognize anxiety and urgency and add emotional information to the search intent.

[1335] 4. Obtain and compare search results

[1336] The server uses a search engine API (e.g., Google Search API) to obtain the current search results for the query.

[1337] The retrieved search results are compared with the generated search intent and sentiment information to determine whether they match.

[1338] 5. Listing

[1339] The server lists queries that do not match the search intent and notifies the administrator of this list, for example by outputting the mismatched queries to a CSV file and sending it by email.

[1340] 6. Generating the Answer Module

[1341] The server generates an answer module that reflects the search intent and sentiment based on the listed queries. It inputs a prompt message to the generative AI model saying, "Please generate a module that provides guidance based on the latest evacuation shelter information," and displays the generated module on a website or app.

[1342] Example scenario

[1343] For example, in an area where an earthquake has occurred, a user searches for "nearby shelter." The server analyzes search logs and extracts the query "nearby shelter." Using a generative AI model, it generates the search intent: "The user is looking for information on the nearest shelter from their current location." At the same time, it uses an emotion engine to recognize the user's anxiety and revise the search intent to "The user is in a hurry and wants to know the specific location of the nearest shelter and how to quickly access it." The server uses a search engine API to obtain current search results for "nearby shelter." If the results are inappropriate, it lists the query and notifies the administrator. Finally, the server generates an answer module that provides the latest shelter location information and posts it on a website or app, allowing users to quickly access it.

[1344] Prompt Sentence Examples

[1345] "Please explain the user's intent for this search query."

[1346] "Determine the sentiment based on the user's input query."

[1347] "Please generate a module that provides guidance based on the latest evacuation shelter information."

[1348] This system can quickly and accurately provide users with the information they need in the event of a disaster. The introduction of an emotion engine makes it possible to respond in accordance with the level of urgency and the user's emotional state, further improving user comfort and convenience.

[1349] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1350] Step 1: Query extraction phase

[1351] Input: Search engine log data

[1352] Processing: The server analyzes the search engine log data and extracts queries related to disasters. Specifically, it downloads the log data using a Python script and uses an NLP library (e.g., NLTK, spaCy) to filter queries containing keywords such as "earthquake occurrence," "evacuation location," and "water outage."

[1353] Output: A list of disaster-related search queries

[1354] Specific behavior:

[1355] The server periodically runs a Python script to retrieve log data.

[1356] The server uses an NLP library to extract queries containing keywords.

[1357] The extracted queries are stored in a database.

[1358] Step 2: Search intent generation phase

[1359] Input: A list of extracted search queries

[1360] Processing: The server uses a generative AI model (e.g., GPT-3) to generate search intent for the search query. It provides the generative AI model with a prompt, such as "Please explain the user's intent for this search query."

[1361] Output: Generated search intent

[1362] Specific behavior:

[1363] The server inputs each search query into a generative AI model.

[1364] The intent obtained from the generative AI model is stored in a database.

[1365] Step 3: Emotion Recognition Phase

[1366] Input: Extracted search query

[1367] Processing: The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotions. Based on the search query entered by the user, it determines emotions such as anxiety or urgency.

[1368] Output: Emotional information (score)

[1369] Specific behavior:

[1370] The server sends the search query to the emotion engine API.

[1371] The idea is to incorporate sentiment scores obtained from the sentiment engine into search intent.

[1372] Step 4: Matching Phase

[1373] Input: Search intent and sentiment information

[1374] Processing: The server uses a search engine API (e.g., Google Search API) to retrieve search results and compare them with the generated search intent and sentiment information to determine if there is a match.

[1375] Output: A list of unmatched search queries

[1376] Specific behavior:

[1377] The server sends the search query to the search engine API and retrieves the results.

[1378] The search results are analyzed and compared with search intent and sentiment information.

[1379] As a result of the comparison, queries that are determined not to match are listed.

[1380] Step 5: List output phase

[1381] Input: A list of unmatched search queries

[1382] Action: The server notifies the administrator of unmatched queries, exports the list as a CSV file, emails it, or displays it in the web dashboard.

[1383] Output: Notification to administrator (CSV file or dashboard)

[1384] Specific behavior:

[1385] The server exports unmatched queries to a CSV file.

[1386] You can either email the administrator or view the list in the web dashboard.

[1387] Step 6: Answer module generation phase

[1388] Input: Unmatched search query and search intent and sentiment information

[1389] Processing: The server generates an answer module based on the unmatched query. For example, it gives a prompt such as "Please generate a module that provides guidance based on the latest evacuation shelter information" to the generative AI model to create appropriate content.

[1390] Output: Answer module

[1391] Specific behavior:

[1392] The server inputs a prompt sentence into the generative AI model and generates an answer module.

[1393] Deploy the generated answer module to your website or app.

[1394] In this way, information based on a user's search query during a disaster can be provided quickly and accurately according to the level of urgency and the user's emotional state.

[1395] (Application example 2)

[1396] 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."

[1397] In the event of a disaster, logistics center employees need to be able to obtain information quickly and appropriately so they can continue their work safely. However, existing systems lack the ability to provide real-time inventory management and delivery information that is appropriate for the unique circumstances of a disaster, and there are also insufficient means to alleviate employee anxiety. This creates problems that make it difficult to operate efficiently.

[1398] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1399] In this invention, the server includes means for extracting search queries relevant to disasters, means for generating search intent for the search queries, means for obtaining search results based on the search queries, means for comparing the search results with the search intent to determine whether they match, means for listing search queries that do not match the search intent, means for recognizing a user's emotion regarding the search queries, means for modifying the search intent based on the emotion, means for providing inventory management and delivery information in the event of a disaster, and means for providing information access using a terminal. This enables employees at a logistics center to obtain necessary information in real time and continue their work efficiently and safely even in the event of a disaster.

[1400] A "search query" is a keyword or phrase that a user enters when searching for information.

[1401] "Search intent" is the specific information or goal a user is seeking when they enter a search query.

[1402] A "search result" is a list of relevant information returned by a search engine in response to a search query.

[1403] "Emotion" refers to the psychological state or feeling a user has when entering a search query.

[1404] "Inventory management" is the activity of understanding the quantity and condition of goods held by logistics centers and companies and managing them appropriately.

[1405] "Shipping information" refers to detailed data and information regarding the shipping of products.

[1406] A "terminal" is a device that is connected to a computer system and inputs and outputs information, such as a smartphone or a head-mounted display.

[1407] MODE FOR CARRYING OUT THE INVENTION

[1408] The present invention is a system that enables employees at a logistics center to quickly and appropriately obtain information and continue their work safely in the event of a disaster, and is composed of the following main components:

[1409] Query Extraction Phase

[1410] First, the server has a means for extracting search queries relevant to disasters. This means analyzes search queries entered by logistics center employees using smartphones or head-mounted displays, and extracts queries that are particularly important in the event of a disaster, such as "delivery interruption," "stock shortage," and "emergency route." These queries are stored in a database.

[1411] Intention generation phase

[1412] The server then uses the generative AI model to generate search intent for each search query stored in the database. For example, for the query "delivery interruption," the server generates the search intent "I'm looking for safe delivery routes and relay points in the event of a disaster." This generated search intent is stored in the database.

[1413] Emotion Recognition Phase

[1414] Furthermore, the server has a means of recognizing employees' emotions using an emotion engine. Based on the search query entered by the user, emotions such as anxiety or urgency are determined and incorporated into the generation of search intent. This allows the search intent to be modified to be more specific and in line with the urgency, for example, "The user wants to know the evacuation route quickly."

[1415] Matching Phase

[1416] Next, the server retrieves the actual search results for each query using the search engine API. It compares the retrieved search results with the generated search intent (including sentiment information) to determine whether they match. For example, if only old route information is displayed for the query "emergency route," it is determined that the query does not match and the query is listed.

[1417] List Output Phase

[1418] The server notifies the administrator of the list of unmatched queries, allowing the administrator to take prompt action, such as providing updated delivery route information or updating inventory status.

[1419] Answer module generation phase

[1420] Finally, the server generates a response module based on the list, providing up-to-date evacuation route information and inventory status, utilizing GIS and real-time data feeds. The response module is displayed on the smartphones and head-mounted displays of logistics center employees for immediate access.

[1421] Specific examples

[1422] Usage Scenarios

[1423] For example, if a logistics center employee types the query "delivery interruption" into their smartphone, the server analyzes the query and generates the search intent: "I'm looking for safe delivery routes and relay points in the event of a disaster." At the same time, the server uses its emotion engine to recognize that the employee is feeling extremely anxious and modifies the search intent to "I want to know about evacuation routes quickly." If the results obtained by the server using the search engine API contain only outdated information, it determines that there is no match and notifies the administrator. An answer module providing the latest delivery route information is then generated and displayed on the employee's device. This process allows logistics center employees to quickly obtain the information they need, even in the event of a disaster, and continue their work efficiently and safely.

[1424] Example of input prompt for generative AI model

[1425] For the query "delivery disruption," generate search intent that reflects a distribution center employee wanting the best delivery route during a disaster.

[1426] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1427] Step 1:

[1428] The server extracts disaster-related search queries by analyzing search logs and filtering queries containing specific keywords (e.g., "delivery interruption," "stock shortage," "emergency route," etc.). The input is the search logs, and the output is a list of disaster-related search queries.

[1429] Step 2:

[1430] The server uses a generative AI model to generate search intent for each extracted search query. For example, for the query "delivery interruption," the server generates the search intent "I'm looking for safe delivery routes and relay points in the event of a disaster." The input is the extracted search query, and the output is the corresponding search intent.

[1431] Step 3:

[1432] The server uses an emotion engine to recognize the emotion of the user (employee) who entered the search query. In this phase, emotions such as anxiety or urgency are determined based on the entered keywords and phrases, and information is added. The input is the search query, and the output is the search intent, including emotional characteristics.

[1433] Step 4:

[1434] The server uses the search engine API to get the current search results based on the revised search intent. Specifically, it calls the search engine API to get the "delivery interruption" information in real time. The input of this step is the revised search intent, and the output is a list of relevant search results.

[1435] Step 5:

[1436] The server compares the retrieved search results with the generated search intent and determines whether they match. For example, it checks whether the search results contain information about safe routes or intermediate stops. The input is the search results and the search intent, and the output is the result of whether they match.

[1437] Step 6:

[1438] The server lists queries that do not match the search intent and notifies the administrator, for example, when up-to-date delivery route information is missing. The input is the unmatched query, and the output is a notification message.

[1439] Step 7:

[1440] The server generates response modules based on the list, providing up-to-date evacuation route information and inventory availability, leveraging a geographic information system (GIS) and real-time data feeds. The input is the latest information or data feed, and the output is the response module.

[1441] Step 8:

[1442] The generated response module is displayed on the user's smartphone or head-mounted display, allowing employees to quickly obtain the information they need in the event of a disaster. The input is the response module, and the output is the information displayed on the device.

[1443] This will enable logistics center employees to continue working efficiently and safely even in the event of a disaster.

[1444] 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.

[1445] 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.

[1446] 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.

[1447] 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.

[1448] 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.

[1449] 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.

[1450] 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).

[1451] 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.

[1452] 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."

[1453] 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.

[1454] 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).

[1455] 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.

[1456] 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.

[1457] 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.

[1458] 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.

[1459] 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.

[1460] 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.

[1461] 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.

[1462] 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.

[1463] 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.

[1464] 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.

[1465] The following is further disclosed regarding the above embodiment.

[1466] (Claim 1)

[1467] A means for extracting search queries related to a disaster;

[1468] means for generating a search intent for the search query;

[1469] means for obtaining search results based on the search query;

[1470] means for comparing the search results with the search intent to determine whether they match;

[1471] means for listing search queries that do not match the search intent;

[1472] A system including:

[1473] (Claim 2)

[1474] means for generating an answer module based on the search intent;

[1475] further comprising means for posting the answer module;

[1476] 10. The system of claim 1.

[1477] (Claim 3)

[1478] The extraction of the search query is performed based on an analysis of a search log.

[1479] 10. The system of claim 1.

[1480] "Example 1"

[1481] (Claim 1)

[1482] A means for extracting search queries related to a disaster;

[1483] means for generating a search intent for the search query;

[1484] means for obtaining search results based on the search query;

[1485] means for comparing the search results with the search intent to determine whether they match;

[1486] means for listing search queries that do not match the search intent;

[1487] means for analyzing whether the search query is searched by many users;

[1488] a means for storing the generated search intent in a database;

[1489] means for notifying an administrator of said list;

[1490] A system including:

[1491] (Claim 2)

[1492] means for generating an answer module based on the search intent;

[1493] means for posting said answer module;

[1494] further comprising means for generating the answer module using a real-time data feed or a geographic information system;

[1495] 10. The system of claim 1.

[1496] (Claim 3)

[1497] The extraction of the search query is performed based on data analysis.

[1498] 10. The system of claim 1.

[1499] "Application Example 1"

[1500] (Claim 1)

[1501] A means for extracting search queries related to a disaster;

[1502] means for generating a search intent for the search query;

[1503] means for obtaining search results based on the search query;

[1504] means for comparing the search results with the search intent to determine whether they match;

[1505] means for listing search queries that do not match the search intent;

[1506] means for obtaining and displaying real-time data based on the listed search queries;

[1507] means for displaying the real-time data on a user terminal;

[1508] A system including:

[1509] (Claim 2)

[1510] means for generating an answer module based on the search intent;

[1511] further comprising means for posting the answer module;

[1512] 10. The system of claim 1.

[1513] (Claim 3)

[1514] The extraction of the search query is performed based on an analysis of a search log.

[1515] 10. The system of claim 1.

[1516] "Example 2: Combining Emotion Engines"

[1517] (Claim 1)

[1518] A means for extracting search queries related to a disaster;

[1519] means for generating a search intent for the search query;

[1520] means for recognizing a user's sentiment from the search query;

[1521] means for incorporating the sentiment based on the search intent;

[1522] means for obtaining search results based on the search query;

[1523] means for comparing the search results with the search intent and the sentiment to determine whether they match;

[1524] means for listing search queries that do not match the search intent;

[1525] A system including:

[1526] (Claim 2)

[1527] The system of claim 1 , further comprising: means for generating and publishing an answer module based on the search intent and the sentiment.

[1528] (Claim 3)

[1529] The system of claim 1 , wherein the extraction of the search queries is performed based on an analysis of a search log.

[1530] "Application example 2 when combining emotion engines"

[1531] (Claim 1)

[1532] A means for extracting search queries related to a disaster;

[1533] means for generating a search intent for the search query;

[1534] means for obtaining search results based on the search query;

[1535] means for comparing the search results with the search intent to determine whether they match;

[1536] means for listing search queries that do not match the search intent;

[1537] means for recognizing a user's sentiment with respect to the search query;

[1538] means for modifying search intent based on the sentiment;

[1539] a means of providing inventory management and delivery information in the event of a failure;

[1540] means for providing information access using a terminal;

[1541] A system including:

[1542] (Claim 2)

[1543] means for generating an appropriate answer module based on the search intent;

[1544] further comprising means for posting the answer module;

[1545] 10. The system of claim 1.

[1546] (Claim 3)

[1547] The extraction of the search query is performed based on an analysis of a search log.

[1548] 10. The system of claim 1. [Explanation of symbols]

[1549] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for extracting search queries related to a disaster; means for generating a search intent for the search query; means for obtaining search results based on the search query; means for comparing the search results with the search intent to determine whether they match; means for listing search queries that do not match the search intent; A system including:

2. means for generating an answer module based on the search intent; further comprising means for posting the answer module; The system of claim 1 .

3. The extraction of the search query is performed based on an analysis of a search log. The system of claim 1 .

Images