system

Abstract

This system aims to improve the satisfaction of residents and tourists, while also stimulating local consumption and increasing the likelihood of visitors staying in the area. [Solution] Means for receiving local information data from local governments and individual stores; means for classifying the received local information data and storing it in a database; means for training an artificial intelligence model to generate recommendations for facilities and stores based on user requests using the stored local information data; and means for receiving requests from users. A system comprising: means for searching a database for relevant information based on a user request and generating recommendations using an artificial intelligence model; means for transmitting the generated recommendation information to the user's terminal; and means for receiving user feedback, storing the received feedback in a database, and using it for retraining.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Although the regional information provided by local governments and individual stores contains information beneficial to many residents and tourists, there is a problem that the information is not used efficiently and effectively. Also, when providing recommendation information for local stores and facilities, the accuracy is low, so in many cases, the results expected by users cannot be obtained. As a result, the activation of consumer activities and the staying power of the region are not sufficiently improved. Therefore, there is a need for a system that efficiently collects regional information and provides highly accurate recommendation information according to user needs.

Means for Solving the Problems

[0005] This invention relates to a system that receives local information data from local governments and individual businesses, classifies that data, and stores it in a database. It also includes means for training an artificial intelligence model to generate recommendations for facilities and businesses based on user requests using the stored local information data. Furthermore, it provides users with optimal recommendation information by receiving user requests, searching for relevant information based on those requests, and generating recommendations using the AI ​​model. It also continuously improves the accuracy of recommendations by receiving user feedback, storing the feedback data in a database, and using it for retraining. Finally, it includes means for updating and deploying the AI ​​model to a production environment, and a user interface using a messaging application. This enables efficient collection and provision of local information, improving the satisfaction of residents and tourists, and stimulating local consumption and increasing their length of stay.

[0006] A "local government" is a public institution that has certain administrative functions within a specific region and provides public services to residents.

[0007] A "privately owned store" refers to a commercial facility or store operated by an individual or a small business.

[0008] "Local information data" refers to data provided by local governments and individual businesses that includes information about facilities, events, shops, and other related matters within a region.

[0009] A "database" is a structured collection of data used to efficiently store, retrieve, and manage information.

[0010] A "user" is an end-user who uses the system to search for information or receive recommendations.

[0011] A "request" is a request message that a user sends to a system to ask for specific information.

[0012] "Recommendation" refers to providing information about facilities and shops that the AI ​​deems optimal based on the user's request.

[0013] An "artificial intelligence model" is an algorithm and mathematical model designed to learn from data and perform a specific task.

[0014] "Feedback" refers to information about evaluations and opinions that users provide after using a facility or store.

[0015] "Retraining" is the process of updating an existing artificial intelligence model based on new data and feedback to improve its accuracy and performance.

[0016] A "user interface" is an interface through which a user and a system exchange information.

[0017] A "messaging application" is a software application that enables the sending and receiving of text and multimedia messages. [Brief explanation of the drawing]

[0018] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Mode for Carrying Out the Invention

[0019] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0020] First, the language used in the following description will be explained.

[0021] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0022] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0023] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0024] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0025] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0026] [First Embodiment]

[0027] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0028] As shown in Figure 1, the 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.

[0029] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0030] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0031] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0032] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0033] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0034] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0036] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0037] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0038] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0039] This invention is a system for collecting local information data from local governments and individual businesses and recommending the most suitable facilities and stores based on user requests. This system operates based on the interaction between a server, a terminal, and a user.

[0040] Collection and storage of local data

[0041] The server receives local information data sent from local governments and individual businesses. The received data is appropriately analyzed and classified into categories such as "facilities," "events," and "restaurants." This classified data is stored in a database. This data forms the basis for generating recommendation information in response to user requests.

[0042] Data preprocessing and AI model training

[0043] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is formatted and converted into a format suitable for machine learning algorithms. This formatted data is used to train the AI ​​model. The newly trained AI model is deployed to the production environment in a timely manner. This enables the provision of highly accurate recommendation information to users.

[0044] Processing user requests

[0045] Users request information about facilities and stores through LINE or other applications. The device sends this request to a server. The server analyzes the received request and searches its database for relevant information. Based on the retrieved information, an AI model generates the most suitable recommendations for the user. These recommendations are then delivered to the user through their device.

[0046] As a concrete example, a user sends a request via the LINE app saying, "Tell me your recommended ramen restaurant." The device sends this request to the server. The server searches its database for information on relevant ramen restaurants and uses an AI model to make the best recommendation. For example, the user might be provided with a recommendation such as, "The ramen restaurant 'Mensho' near the station has high ratings."

[0047] Gathering feedback and relearning

[0048] After a user uses a recommended facility or store, they provide feedback. The device sends this feedback to the server. The server stores the feedback data in a database and uses it for retraining. This new feedback data is also integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[0049] As a concrete example, a user sends feedback saying "It was very delicious" about a ramen restaurant they visited. This feedback is sent to the server and stored in the database. The server uses this feedback data to retrain the AI ​​model and incorporate it into future recommendations.

[0050] Thus, the present invention is a system that efficiently collects local information from local governments and individual shops and provides users with highly accurate recommendation information, thereby stimulating local consumer activity and improving the likelihood of people staying in the area.

[0051] The following describes the processing flow.

[0052] Collection and storage of local data

[0053] Step 1:

[0054] The server receives HTTP POST requests from local governments and individual businesses.

[0055] Step 2:

[0056] The server parses the received request body and extracts regional information data.

[0057] Step 3:

[0058] The server categorizes the extracted data into categories such as "facilities," "events," and "restaurants."

[0059] Step 4:

[0060] The server stores the classified data in a database.

[0061] Data preprocessing and AI model training

[0062] Step 1:

[0063] The server periodically checks the raw data in the database, corrects incomplete data, and removes noisy data.

[0064] Step 2:

[0065] The server formats the cleansed data into a format suitable for machine learning algorithms.

[0066] Step 3:

[0067] The server uses the formatted data to train the AI ​​model.

[0068] Step 4:

[0069] The server evaluates the model's performance and adjusts parameters as needed.

[0070] Step 5:

[0071] The server uploads the newly trained model to the production deployment environment.

[0072] Processing user requests

[0073] Step 1:

[0074] Users request information about facilities and stores through LINE or other apps.

[0075] Step 2:

[0076] The terminal sends the user's request to the server as an HTTP request.

[0077] Step 3:

[0078] The server parses the received request and understands its contents.

[0079] Step 4:

[0080] The server searches the database for information related to the request.

[0081] Step 5:

[0082] The server uses an AI model to generate optimal recommendations based on the search results.

[0083] Step 6:

[0084] The server formats the generated recommendation information and sends it to the terminal.

[0085] Step 7:

[0086] The device displays recommendation information on the user interface.

[0087] Gathering feedback and relearning

[0088] Step 1:

[0089] Users provide feedback after using the recommended facilities or stores.

[0090] Step 2:

[0091] The device sends feedback to the server.

[0092] Step 3:

[0093] The server saves the feedback to the database.

[0094] Step 4:

[0095] The server integrates the new feedback data with the existing data.

[0096] Step 5:

[0097] The server retrains the AI ​​model using the integrated data.

[0098] In this way, the system achieves highly accurate information delivery through efficient processes of collecting, storing, analyzing, learning from, recommending, and providing feedback on local information.

[0099] (Example 1)

[0100] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0101] Conventional local information recommendation systems often suffered from incomplete data collection or reduced recommendation accuracy due to the inclusion of noisy data. Furthermore, responses to user requests were sometimes not in real time, compromising usability. In addition, insufficient retraining using feedback made continuous improvement of recommendation accuracy difficult. This invention aims to effectively solve these problems and improve the recommendation accuracy for local facilities and stores.

[0102] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0103] In this invention, the server includes means for receiving local information data from local governments and individual stores; means for analyzing the received local information data, classifying it into categories such as "facilities," "events," and "restaurants," and storing it in a database; means for periodically cleansing the stored local information data to correct defects and remove noise data; means for training an artificial intelligence model using a machine learning algorithm with the cleansed data; means for deploying the trained artificial intelligence model; means for analyzing user requests using natural language processing technology; means for searching for relevant information from the database based on the analyzed requests and generating recommendation information using the artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; and means for receiving user feedback, storing the received feedback in a database, and using it for retraining. This improves the accuracy of local information data collection and cleansing, and enables real-time responses. Furthermore, the recommendation accuracy continuously improves by incorporating feedback.

[0104] "Local information data" refers to various types of information provided by local governments and individual businesses, such as facility names, addresses, contact information, and event information.

[0105] "Analysis" refers to the process of appropriately assigning and classifying received data into each category.

[0106] A "category" refers to a specific group used to classify information such as "facilities," "events," and "restaurants."

[0107] A "database" refers to a data management system for efficiently storing, searching, and updating information.

[0108] "Cleansing" refers to the process of correcting data defects and removing noisy data.

[0109] A "machine learning algorithm" refers to a mathematical model that uses large amounts of data to detect patterns and makes predictions and classifications based on new data.

[0110] An "artificial intelligence model" refers to a system that uses machine learning algorithms to analyze data, make predictions, and generate recommendation information.

[0111] "Deployment" refers to introducing a trained artificial intelligence model into a production environment and putting it into operation.

[0112] "Natural language processing technology" refers to computer science techniques used to analyze human language and understand its meaning.

[0113] "Recommendation information" refers to information about the most suitable facilities and stores, generated based on user requests.

[0114] "Feedback" refers to information such as ratings and comments that users provide after using a recommended facility or store.

[0115] "Retraining" refers to the process of retraining an artificial intelligence model using newly collected data and feedback data to improve the model's accuracy and performance.

[0116] This invention is a system that collects local information data from local governments and individual businesses and recommends the most suitable facilities and businesses based on user requests. This system operates based on the interaction of a server, terminals, and users.

[0117] Collection and storage of local data

[0118] The server receives local information data sent from local governments and individual businesses. This data includes facility names, addresses, contact information, and event information. The received data is analyzed by the server and classified into categories such as "facilities," "events," and "restaurants." The classified data is stored in a database (for example, MySQL® or PostgreSQL). This data forms the basis for generating recommendation information in response to user requests.

[0119] Data preprocessing and AI model training

[0120] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is converted into a format suitable for machine learning algorithms. For example, the Google® Maps API is used to verify the accuracy of addresses to compensate for data defects. Using the formatted data, the server trains an AI model (e.g., using TENSORFLOW® or PyTorch). After training is complete, the newly trained AI model is deployed to the production environment.

[0121] Processing user requests

[0122] Users request information about specific facilities or stores via LINE or a dedicated app. For example, they might send a request like, "Tell me your recommended ramen restaurant." The device sends this request to the server in real time. The server analyzes the received request using natural language processing technology (e.g., Google NLP API or SpaCy). It then searches the database for relevant information, and an AI model generates the most suitable recommendation information.

[0123] Providing recommendation information

[0124] The server sends the generated recommendation information to the device. For example, the user is provided with a recommendation such as, "The ramen shop in front of the station has high ratings." The device then displays this information to the user. In the case of LINE, it is displayed as a message, and in the case of a dedicated app, it is displayed as an in-app notification or on the screen.

[0125] Gathering feedback and relearning

[0126] Users provide feedback after using recommended facilities or stores. For example, they might send a simple rating such as "It was delicious." The device sends this feedback to the server. The server stores the feedback data in a database and uses it for training the next AI model. The AI ​​model, retrained with the new feedback data, generates even more accurate recommendations.

[0127] Specific example

[0128] A user sends a request via the LINE app saying, "Tell me your recommended ramen restaurant." The device sends this request to the server. The server searches its database for information on relevant ramen restaurants and uses an AI model to make the best recommendation. For example, the user is provided with a recommendation such as, "The ramen restaurant '○○' near the station has high ratings."

[0129] By presenting specific situations in this way, the generative AI model becomes more likely to generate specific and appropriate flows.

[0130] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0131] Step 1: Collection and reception of local information data

[0132] The server receives local information data sent from local governments and individual businesses via APIs and database connections. Specifically, the server uses API crawling and scraping techniques to retrieve data such as facility names, addresses, contact information, and event information. The input is data from local governments and individual businesses, and the output is raw local information data stored in the server's internal database.

[0133] Step 2: Analysis and classification of regional information data

[0134] The server analyzes the received data using natural language processing techniques (e.g., Google NLP API or SpaCy). Specifically, it performs text analysis and keyword matching to classify the data into categories such as "facilities," "events," and "restaurants." The input is the raw data received in step 1, and the output is the data categorized.

[0135] Step 3: Save data

[0136] The server stores the classified data in a relational database (e.g., MySQL or PostgreSQL). Specifically, the server uses SQL queries to store the data in the appropriate tables within the database. The input is the data classified in step 2, and the output is the structured data stored in the database.

[0137] Step 4: Data Cleansing

[0138] The server periodically checks the data in the database, corrects incomplete data, and removes noisy data. Specifically, it uses the Google Maps API to fill in missing address information and standardizes data with inconsistent formats. The input is existing data in the database, and the output is cleansed and formatted data.

[0139] Step 5: Training the AI ​​model

[0140] The server uses the cleansed data to train an AI model using machine learning algorithms (e.g., TensorFlow or PyTorch). Specifically, it processes large amounts of data on a GPU (e.g., NVIDIA Tesla V100) and learns patterns to generate the expected output. The input is the cleansed data, and the output is the trained AI model.

[0141] Step 6: Deploying the Artificial Intelligence Model

[0142] The server deploys the trained AI model to the production environment. Specifically, it uploads the model to the server and makes it accessible via API. The input is the trained AI model, and the output is the publicly available AI model.

[0143] Step 7: Receiving and parsing user requests

[0144] Users request information about specific facilities or stores via LINE or a dedicated application. The device sends this request to the server in real time. The server analyzes the received request using natural language processing technology and converts it into structured data. The input is the user request, and the output is the analyzed request data.

[0145] Step 8: Generating and searching for recommendation information

[0146] The server searches the database for relevant information based on the parsed request and generates optimal recommendations using an AI model. Specifically, the server uses collaborative filtering and content-based filtering to recommend the most suitable facilities and stores to the user. The input is the parsed request data and relevant data in the database, and the output is the recommendations.

[0147] Step 9: Submitting and displaying recommendation information

[0148] The server sends the generated recommendation information to the device. The device then displays this information to the user. Specifically, in the case of LINE, it is provided to the user as a message, and in the case of a dedicated app, it is provided as an in-app notification or screen display. The input is the recommendation information, and the output is the recommendation content provided to the user.

[0149] Step 10: Gathering Feedback and Retraining

[0150] Users provide feedback after using recommended facilities or stores. Specifically, users send ratings and comments via a dedicated app or LINE. The device sends this feedback to a server. The server stores the feedback data in a database and retrains the AI ​​model by incorporating the new data. The input is the user's feedback, and the output is the retrained AI model.

[0151] (Application Example 1)

[0152] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0153] Traditional facility and store recommendation systems suffer from insufficient collection and analysis of local information, making it difficult to respond quickly and appropriately to user requests. Furthermore, the inability to provide personalized recommendations based on user preferences and usage history makes improving user satisfaction a challenge. Additionally, the inability to offer value-added services such as real-time event notifications makes it difficult to maintain user interest.

[0154] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0155] In this invention, the server includes means for receiving local information data from local governments and individual stores; means for classifying the received local information data and storing it in a database; means for training an artificial intelligence model to generate facility and store recommendations based on user requests using the stored local information data; means for receiving requests from users; means for searching the database for relevant information based on user requests and generating recommendations using the artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; means for receiving feedback from users, storing the received feedback in a database and using it for retraining; means for processing user requests via an application installed on a smartphone; and means for providing information on recommended stores and real-time event notifications via push notifications. This enables personalized recommendations of facilities and stores tailored to the user's interests and preferences, and improves the user experience through real-time information provision and feedback reflection.

[0156] A "local government" is a public institution that carries out administrative duties in a specific region and is a source of regional information data.

[0157] A "personal store" is a small-scale commercial facility or service provider owned or operated by an individual, and is a source of local information data.

[0158] "Local information data" refers to information about facilities, shops, events, and services located within a specific area, and is provided by local governments and individual businesses.

[0159] A "database" is a digital system for organizing and storing regional information data in a searchable and accessible format.

[0160] An "artificial intelligence model" is an algorithm or computational model that learns from a large amount of data and performs a specific task; in this context, it refers to a model used in recommendation systems.

[0161] A "user terminal" refers to a device used by a user, such as a computer, smartphone, or tablet, which is used to receive recommendation information and send requests.

[0162] "Feedback" refers to opinions and impressions, such as ratings and comments, provided by users, and is information used for system improvement and retraining.

[0163] "Retraining" is the process of updating an existing artificial intelligence model using newly collected data and feedback to improve its performance and accuracy.

[0164] A "smartphone" is a portable computer with mobile communication and internet connectivity capabilities, and is a device that allows for the installation and use of applications.

[0165] "Push notifications" are notification messages automatically sent from a server to a user's device, providing a means of delivering important information and updates in real time.

[0166] A "user interface" is an interface through which a user interacts with a system, providing information using messaging applications and push notification functions.

[0167] To implement this invention, interaction between a server, a terminal, and a user is necessary. A detailed embodiment is shown below.

[0168] Collection and storage of local data

[0169] First, the server receives local information data from local governments and individual businesses. This data is categorized into categories such as "facilities," "events," and "restaurants." The categorized data is stored in a database such as Firebase. This data forms the basis for generating recommendation information in response to user requests.

[0170] Data preprocessing and AI model training

[0171] Next, the server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is then formatted and converted into a format suitable for machine learning algorithms. Using this formatted data, an AI model is trained using Python libraries such as Scikit-learn. The newly trained AI model is deployed to the production environment as needed. This enables the provision of highly accurate recommendation information to users.

[0172] Processing user requests

[0173] Users can request information about facilities and shops through applications or messaging apps installed on their smartphones. In this case, if a user sends a request such as "Tell me a good ramen shop," the device sends this request to the server. The server analyzes the received request and searches its database for relevant information. Based on the retrieved information, an AI model generates the most suitable recommendations for the user. The generated recommendations are then sent as push notifications to the user's smartphone, for example, "The ramen shop in front of the station has high ratings."

[0174] Gathering feedback and relearning

[0175] After a user visits a recommended facility or store, the terminal sends feedback to the server. For example, a comment such as "It was delicious" might be sent. The server stores this feedback data in a database and uses it for retraining. This new feedback data is also integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[0176] Specific example

[0177] As a concrete example, when a user sends a request using a smartphone application saying, "Tell me some recommended cafes," the server searches for relevant cafes in real time, and the AI ​​model can recommend cafes with high ratings. The user is notified of the recommendation, such as, "The cafe near the station has high ratings." After the user visits the cafe, they can send feedback such as, "It was very comfortable," and this information will be used to train the AI ​​model for the next time.

[0178] This system will allow users to receive recommendations for facilities and shops that best suit their interests and preferences, and is expected to further stimulate local consumer activity.

[0179] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0180] Step 1:

[0181] The server receives local information data from local governments and individual businesses and stores it in a database. Specifically, it retrieves data via CSV files or APIs and stores it in the database as structured data. The input is local information data, and the output is organized and categorized data.

[0182] Step 2:

[0183] The server periodically checks the regional information data in the database, corrects incomplete data, and removes noisy data. For example, it imputes missing values ​​and removes inappropriate values. This process results in cleansed data. The input is the regional information data in the database, and the output is the cleansed data.

[0184] Step 3:

[0185] The server uses the cleansed data to transform it into a format suitable for machine learning algorithms and trains an AI model. It uses Python's Scikit-learn to vectorize the data and train the model. The input is the cleansed data, and the output is the trained AI model.

[0186] Step 4:

[0187] Users request information about facilities and stores through a smartphone application. For example, they might type "Tell me your recommended ramen restaurant" as text. The input is the user's request, and the output is the request data.

[0188] Step 5:

[0189] The terminal sends the user's request to the server. The server parses the request and searches the database for relevant information. For example, it might perform a search based on the keyword "ramen shop." The input is the request data, and the output is the relevant information.

[0190] Step 6:

[0191] The server generates optimal recommendations using an AI model based on the searched information. It uses a trained model to select highly-rated stores. The input is relevant information, and the output is recommendations.

[0192] Step 7:

[0193] The server sends the generated recommendation information to the user's terminal. For example, it might send a push notification saying, "The ramen shop in front of the station has high ratings." The input is the recommendation information, and the output is the notification sent to the user's terminal.

[0194] Step 8:

[0195] Users visit recommended facilities or shops and then provide feedback. They submit comments such as "It was delicious." The input is the user's feedback, and the output is feedback data.

[0196] Step 9:

[0197] The device sends user feedback to the server. The server stores the feedback data in a database and uses it for retraining. The collected feedback data is also integrated with existing data and used to retrain the AI ​​model. The input is the feedback data, and the output is the updated AI model.

[0198] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0199] This invention is a system for collecting local information data from local governments and individual businesses, recognizing user emotions using an emotion engine based on user requests and feedback, and utilizing that information to recommend the most suitable facilities and businesses. This system operates based on the interaction between a server, a terminal, and a user.

[0200] Collection and storage of local data

[0201] The server receives local information data sent from local governments and individual businesses. The received data is appropriately analyzed, categorized into categories such as "facilities," "events," and "restaurants," and stored in a database. This data forms the basis for generating recommendation information in response to user requests.

[0202] Data preprocessing and AI model training

[0203] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is formatted and converted into a format suitable for machine learning algorithms. This formatted data is used to train the AI ​​model. The newly trained AI model is deployed to the production environment in a timely manner. This enables the provision of highly accurate recommendation information to users.

[0204] Processing user requests and sentiment analysis

[0205] Users request information about facilities and stores through LINE or other applications. The device sends this request to the server. The server analyzes the received request and understands its content. Furthermore, it can use an emotion engine to extract emotional information from the user's request. For example, if a user requests "I'm hungry and very irritated," the emotion engine recognizes "irritated."

[0206] Generation and provision of recommendation information

[0207] The server searches the database for information related to the request. Furthermore, using the retrieved sentiment information, an AI model generates recommendations best suited to the user. For example, a user who is "frustrated" might be recommended a restaurant that can provide service quickly. This recommendation information is then delivered to the user via their device.

[0208] As a concrete example, a user requests via the LINE app, "Tell me about a good ramen restaurant quickly." The device sends this request to the server. The server analyzes the request and uses an emotion engine to obtain "urgent" emotion information. The server searches its database for ramen restaurants that can respond quickly and uses an AI model to recommend, "The ramen restaurant 'Mensho' near the station is highly rated for its quick service." The device then provides this information to the user.

[0209] Gathering feedback and relearning

[0210] After a user visits a recommended facility or store, they provide feedback. The device sends this feedback to a server. The server stores the feedback data in a database and uses it for retraining. The sentiment information included in the provided feedback is also analyzed and stored in the database. This new feedback data is integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[0211] As a concrete example, a user might submit feedback about a ramen restaurant they visited, stating, "It was very delicious, but the service was a little slow." This feedback is sent to the server and stored in the database. The server uses this feedback data and the sentiment information analyzed by the sentiment engine to retrain the AI ​​model and incorporate it into future recommendations.

[0212] Thus, the present invention is a system that efficiently collects local information from local governments and individual shops, analyzes users' emotions using an emotion engine, and provides users with highly accurate recommendation information, thereby stimulating local consumer activity and improving their dwell time.

[0213] The following describes the processing flow.

[0214] Collection and storage of local data

[0215] Step 1:

[0216] The server receives HTTP POST requests from local governments and individual businesses.

[0217] Step 2:

[0218] The server parses the received request body and extracts regional information data.

[0219] Step 3:

[0220] The server categorizes the extracted data into categories such as "facilities," "events," and "restaurants."

[0221] Step 4:

[0222] The server stores the classified data in a database.

[0223] Data preprocessing and AI model training

[0224] Step 1:

[0225] The server periodically checks the raw data in the database, corrects incomplete data, and removes noisy data.

[0226] Step 2:

[0227] The server formats the cleansed data into a format suitable for machine learning algorithms.

[0228] Step 3:

[0229] The server uses the formatted data to train the AI ​​model.

[0230] Step 4:

[0231] The server evaluates the model's performance and adjusts parameters as needed.

[0232] Step 5:

[0233] The server uploads the newly trained model to the production deployment environment.

[0234] Processing user requests and sentiment analysis

[0235] Step 1:

[0236] Users request information about facilities and stores through LINE or other apps.

[0237] Step 2:

[0238] The terminal sends the user's request to the server as an HTTP request.

[0239] Step 3:

[0240] The server analyzes the received request to understand its content and the user's intent.

[0241] Step 4:

[0242] The server uses an emotion engine to extract emotional information from user requests.

[0243] Step 5:

[0244] The server searches the database for information related to the request.

[0245] Step 6:

[0246] The server uses an AI model to generate optimal recommendations based on search results and sentiment information.

[0247] Step 7:

[0248] The server formats the generated recommendation information and sends it to the terminal.

[0249] Step 8:

[0250] The device displays recommendation information on the user interface.

[0251] Specific example

[0252] Step 1:

[0253] The user requests "Tell me a delicious ramen shop quickly" on the LINE app.

[0254] Step 2:

[0255] The terminal sends this request to the server as an HTTP request.

[0256] Step 3:

[0257] The server analyzes the request and understands that the user is "in a hurry".

[0258] Step 4:

[0259] The server uses the emotion engine to extract the "in a hurry" emotion from the text of the request.

[0260] Step 5:

[0261] The server quickly searches the database for information on ramen shops that can provide services.

[0262] Step 6:

[0263] Based on the search results and emotion information, the server generates optimal recommendation information using an AI model.

[0264] Step 7:

[0265] The server formalizes the recommendation information "The ramen shop 'Noodle Master' in front of the station is evaluated for quick service" and sends it to the terminal.

[0266] Step 8:

[0267] The terminal displays this information to the user on the LINE app.

[0268] Collection of feedback and re - learning

[0269] Step 1:

[0270] Users provide feedback after using the recommended facilities or stores.

[0271] Step 2:

[0272] The device sends feedback to the server.

[0273] Step 3:

[0274] The server analyzes the feedback and extracts emotional information.

[0275] Step 4:

[0276] The server stores feedback data and emotional information in a database.

[0277] Step 5:

[0278] The server retrains the AI ​​model using the stored feedback data.

[0279] In this way, the system achieves highly accurate information provision and recommendations tailored to user sentiment through an efficient process of collecting, storing, analyzing, learning, recommending, and providing feedback on local information.

[0280] (Example 2)

[0281] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0282] Current regional information recommendation systems have a problem in that they struggle to make recommendations that take into account user emotions and urgency. Furthermore, the presence of data incompleteness and noise often leads to decreased recommendation accuracy. Additionally, there is a lack of effective means to utilize user feedback for retraining, which hinders the continuous improvement of system performance.

[0283] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in Example 2 is realized by the following means.

[0284] In this invention, the server includes means for receiving regional information data from local governments and individual stores, means for classifying the received regional information data and storing it in a database, means for checking the consistency of the stored regional information data and correcting and deleting incomplete data and noise data, means for converting the formatted data into a form suitable for a machine learning algorithm and training an artificial intelligence model, means for receiving a request from a user, means for analyzing the content of the request and obtaining emotion information using an emotion engine, means for searching for relevant information from a database and generating recommendation information using an artificial intelligence model, means for transmitting the generated recommendation information to a user terminal, means for receiving feedback from a user, storing the received feedback in a database and using it for re-learning, and means for optimizing the recommendation information based on the generated emotion information. Thereby, it becomes possible to provide highly accurate recommendation information based on the emotions and urgency of the user.

[0285] "Local government" refers to a public institution for managing and operating a specific region.

[0286] "Individual store" refers to a commercial facility or service-providing facility operated by an individual or a small-scale business operator.

[0287] "Regional information data" refers to data including information on facilities, events, restaurants, etc. related to a specific region.

[0288] "Database" refers to a system for systematically storing and managing collected information.

[0289] "Incomplete data" refers to data lacking necessary information or containing errors.

[0290] "Noise data" refers to unwanted data that hinders analysis and learning.

[0291] A "machine learning algorithm" refers to a method or model for automatically learning patterns and rules from data.

[0292] An "artificial intelligence model" refers to a program that has been trained using machine learning algorithms and possesses the ability to perform specific tasks.

[0293] An "emotion engine" refers to a technology that uses natural language processing and text mining to recognize and analyze emotions from text.

[0294] A "user terminal" refers to an electronic device (e.g., smartphone, tablet, personal computer) used by a user to input and receive information.

[0295] A "request" refers to a request made by a user to a system for the provision of information.

[0296] "Feedback" refers to the evaluations and opinions that users give regarding the information and services provided.

[0297] "Retraining" refers to the process of retraining an artificial intelligence model using new data and feedback.

[0298] "Optimization" refers to adjusting the parameters and processes of a system to their optimal state in order to achieve a specific objective.

[0299] This invention relates to a system that collects local information data from local governments and individual businesses and generates recommendation information based on user requests and feedback. This system operates based on the interaction between a server, a terminal, and a user.

[0300] Collection and storage of local data

[0301] The server receives local information data sent from local governments and individual businesses. This information is retrieved through interfaces such as APIs. Examples include event information from local governments and business hours from individual businesses. This data is categorized into categories such as "facilities," "events," and "restaurants," and stored in a database. Relational databases such as MySQL and PostgreSQL can be used as the database. This data later forms the basis for generating recommendation information in response to user requests.

[0302] Data preprocessing and AI model training

[0303] The server periodically checks the received regional data, correcting and removing incomplete and noisy data. Specifically, it fills in missing information and removes outliers. Data cleansing scripts using Python or Pandas are used. The formatted data is converted into a format suitable for machine learning algorithms. For example, categorical data is encoded into numerical data. This is then used to train the AI ​​model. TensorFlow and PyTorch are used as machine learning frameworks. The newly trained AI model is deployed to the production environment as needed.

[0304] Processing user requests and sentiment analysis

[0305] Users request information about facilities and stores through LINE or a dedicated application. For example, they might send a message like, "Tell me about a good ramen restaurant quickly." The device sends this request to the server. The server analyzes the received request and uses natural language processing technology (e.g., spaCy or NLTK) to understand its content. Furthermore, it uses an emotion engine (e.g., Google Cloud Natural Language API) to extract emotional information from the user's request. For example, it recognizes "irritated" from a request like, "I'm hungry and very irritated."

[0306] Generation and provision of recommendation information

[0307] The server searches the database for information related to the request, and the AI model uses the obtained sentiment information to generate optimal recommendation information for the user. For example, recommend a restaurant that can provide services quickly to a "frustrated" user. This recommendation information is provided to the user through the terminal. As a specific example, when the user requests "Tell me a delicious ramen restaurant quickly" on the LINE app, the terminal sends the request to the server, and the server analyzes the request and sentiment information. Search for ramen restaurant information that can respond quickly from the database, and use the AI model to recommend "The ramen restaurant '〇〇' in front of the station is evaluated for quick service". This information is provided to the user through the terminal.

[0308] Collection and retraining of feedback

[0309] The user provides feedback after using the recommended facilities or stores. For example, send feedback such as "It was very delicious, but the service was a bit slow" on the LINE app. The terminal sends this feedback to the server. The server saves the received feedback data in the database and uses it for retraining. The collected feedback data goes through cleansing and formatting steps and is integrated with the existing data. Then it is used for retraining the AI model. This enables the system to always provide highly accurate recommendation information.

[0310] Specific examples of prompt sentences

[0311] The following are specific examples of input prompt sentences for the generation AI model.

[0312] 1. "Tell me a delicious café nearby"

[0313] 2. "Looking for a restaurant where I can feel safe even with children"

[0314] 3. "What are some recommended events to go to with friends?"

[0315] 4. "Please recommend a good place for a date."

[0316] By utilizing these prompts, the generative AI model can provide optimal recommendation information based on user requests.

[0317] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0318] Step 1: Collecting local data

[0319] The server receives local information data from local governments and individual businesses. The specific operation is as follows: First, input processing is performed to acquire data through an interface such as an API. Next, the received data is temporarily stored in a buffer. Then, it is classified into categories such as "facilities," "events," and "restaurants," and finally stored in the database. In this step, the input is the received local information data, and the output is the classified and stored data.

[0320] Step 2: Data preprocessing

[0321] The server periodically checks the received regional information data, correcting and removing incomplete and noisy data. Specifically, it performs data integrity checks, supplements missing information, and removes outliers. Next, it formats this data into a format suitable for machine learning algorithms. For example, it encodes categorical data into numerical data. In this step, the input is classified data, and the output is formatted data.

[0322] Step 3: Training the AI ​​model

[0323] The server trains an AI model using pre-processed data. The specific operation is as follows: First, the new data is split into a training set and a test set. Next, the AI ​​model is trained using a machine learning framework such as TensorFlow or PyTorch. The trained model is finally deployed to the production environment. The input in this step is a formatted dataset, and the output is the trained AI model.

[0324] Step 4: Processing User Requests

[0325] Users send requests via LINE or a dedicated application. A concrete example of such a request might be, "Tell me a good ramen restaurant quickly." The device sends this request to the server. The server uses natural language processing techniques to analyze the received request and understand its content. In this step, the input is the user request, and the output is the analyzed request content.

[0326] Step 5: Emotion Analysis

[0327] The server analyzes the received request using an emotion engine to obtain the user's emotion information. Specifically, it uses a natural language processing library to extract keywords and emotional expressions from the request and obtains the emotion information through the emotion engine. For example, from the request "Tell me a good ramen restaurant quickly," the emotion information "I'm in a hurry" is extracted. In this step, the input is the analyzed request content, and the output is the obtained emotion information.

[0328] Step 6: Generating recommendation information

[0329] The server searches the database for relevant information and generates optimal recommendations based on the retrieved sentiment information and user requests. Specifically, it uses an AI model to evaluate the request content and sentiment data to generate optimal recommendations. For example, it might recommend stores that can provide quick service to a user who is "in a hurry." The inputs in this step are sentiment information and user requests, and the output is the generated recommendations.

[0330] Step 7: Providing Recommendation Information

[0331] The server provides the generated recommendation information to the user via the terminal. Specifically, the recommendation information is sent to the terminal via an API, and the terminal displays that information to the user. For example, a recommendation such as "The ramen shop 'XX' in front of the station is highly rated for its fast service" is notified to the LINE app. In this step, the input is the generated recommendation information, and the output is the notification to the user.

[0332] Step 8: Gathering Feedback

[0333] Users provide feedback after using a facility or store. For example, they might say, "The food was delicious, but the service was a little slow." The device sends this feedback to the server, which then stores it in a database. In this step, the input is the user feedback, and the output is the stored feedback data.

[0334] Step 9: Retraining the AI ​​model

[0335] The server retrains the AI ​​model using the collected feedback data. Specifically, it cleanses and formats the new feedback data and integrates it with the existing dataset. Then, it retrains the AI ​​model and deploys it to the production environment. In this step, the input is the saved feedback data, and the output is the retrained AI model.

[0336] (Application Example 2)

[0337] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0338] Traditional local information recommendation systems have a problem of not providing sufficient user satisfaction because they simply provide information without considering the user's emotional state. Furthermore, online virtual shopping lacks personalized recommendations based on user emotions and feedback. This results in a uniform shopping experience for all users, making it difficult to recommend optimal products and services that meet individual needs.

[0339] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for receiving local information data from local governments and individual stores; means for classifying the received local information data and storing it in a database; means for training an artificial intelligence model for generating facility and store recommendations based on user requests using the stored local information data; means for searching for relevant information from the database based on user requests and sentiment information and generating recommendations using the artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; means for receiving user feedback, storing the received feedback in a database and using it for retraining; and means for recommending products and services in the virtual store based on the stored local information data and user sentiment information. This makes it possible to provide detailed information recommendations based on user sentiment and feedback, thereby improving user satisfaction and the shopping experience.

[0340] A "local government" is an administrative body that governs a specific region and is responsible for providing public services and promoting the development of the local community.

[0341] A "private store" is a commercial facility owned and operated by an individual, primarily providing goods or services.

[0342] "Local information data" refers to data provided by local governments or individual businesses that includes various types of information about a particular region.

[0343] A "database" is a system for efficiently storing, searching, and managing data.

[0344] "Emotional information" refers to data that indicates a user's emotional state, and is primarily analyzed by an emotion engine.

[0345] A "user terminal" is a device used by a user to interact with the system, and includes smartphones, tablets, and other similar devices.

[0346] An "artificial intelligence model" is an algorithm or system that automatically performs a specific task by analyzing and learning from data.

[0347] "Feedback" refers to data, including user opinions and impressions after using a system, which is used to improve the system.

[0348] A "virtual store" is a commercial facility in an online environment that provides goods and services via the internet.

[0349] "Recommendation information" refers to information about the most suitable facilities and stores, generated based on user requests and sentiments.

[0350] "Retraining" is the process of retraining an existing artificial intelligence model using new data and feedback.

[0351] This invention is a system that collects local information data and generates and provides recommendation information based on user requests and sentiment information. Specific embodiments of this invention are described below.

[0352] 1. Data collection and storage

[0353] The server receives local information data from local governments and individual businesses. This data includes a wide range of information such as facilities, events, and restaurants. The received data is appropriately analyzed, categorized, and then stored in a database.

[0354] 2. Data preprocessing and AI model training

[0355] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is formatted to a format suitable for machine learning algorithms, and this is used to train AI models. The newly trained AI models are deployed to the production environment, enabling them to provide users with highly accurate recommendations.

[0356] 3. Processing User Requests and Sentiment Analysis

[0357] Users request information through the virtual store's shopping app or messaging application. The device sends this request to the server. The server analyzes the received request and uses an emotion engine to obtain the user's emotional information. For example, if a user requests, "I'm feeling stressed today, so I'm looking for a relaxing cafe. Do you have any recommendations?", the emotion engine recognizes "stress."

[0358] 4. Generation and provision of recommendation information

[0359] The server searches the database for information related to the request, and using the retrieved sentiment information, the AI ​​model generates recommendations best suited to the user. For example, a user feeling "stressed" would be recommended relaxing cafes or shops. This recommendation information is then sent to the user's device.

[0360] 5. Gathering feedback and relearning

[0361] After a user uses a recommended facility or store, they provide feedback. The device sends this feedback to a server. The server stores the feedback data in a database and uses it for retraining. The sentiment information included in the provided feedback is also analyzed and stored in the database. This new feedback data is integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[0362] Hardware and software to be used

[0363] Hardware: Servers, smartphones, tablets

[0364] Software: Web frameworks (e.g., Flask, Django), machine learning libraries (e.g., TensorFlow), sentiment analysis engines (e.g., Emotion-Recognition-API)

[0365] Specific example

[0366] When a user requests, "I'm feeling stressed today, so I'm looking for a relaxing cafe. Do you have any recommendations?", the emotion engine recognizes the user's "stress" and recommends a relaxing establishment. Similarly, if a user requests, "Tell me about a good ramen shop quickly," the system can recommend a ramen shop that can provide quick service based on the emotional information that the user is "in a hurry."

[0367] Example of a prompt:

[0368] "I'm feeling stressed today, so I'm looking for a relaxing cafe. Do you have any recommendations?"

[0369] "Please tell me a good ramen shop quickly. Thank you."

[0370] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0371] Step 1:

[0372] The server receives local information data from local governments and individual businesses. This data is often provided in a structured format such as JSON. The received data is saved to a database in real time. The input is local information data sent from local governments and individual businesses, and the output is the saved data.

[0373] Step 2:

[0374] The server classifies the data in the database and stores it by category. For example, it might classify data into categories such as "restaurants," "events," and "facilities." This classification is performed using a data analysis algorithm. The input is the stored regional information data, and the output is the classified data.

[0375] Step 3:

[0376] The server uses the cleansed data to train an artificial intelligence model. Data shaping processes convert the data into a format suitable for machine learning algorithms, and this is then input into the AI ​​model. This dataset is used to train the model and build a highly accurate recommendation engine. The input is the shaped data, and the output is the trained AI model.

[0377] Step 4:

[0378] The user sends a request via a smartphone or tablet. The request content is parsed using natural language processing. The device sends this request to the server, and the prompt text includes the user's sentiment. The input is the user request, and the output is the parsing result.

[0379] Step 5:

[0380] The server analyzes the received request and uses an emotion engine to obtain the user's emotional information. For example, the emotion engine analyzes emotional information such as "I want to relax," "I'm in a hurry," or "I'm stressed." The input is the user's request, and the output is the user's emotional information.

[0381] Step 6:

[0382] The server searches the database for relevant information and uses an AI model to generate optimal recommendations based on the retrieved sentiment information. For example, based on sentiment information such as "I want to relax," it might recommend quiet cafes or parks. The input is the user's sentiment information and request, and the output is the recommendations.

[0383] Step 7:

[0384] The server sends the generated recommendation information to the user's terminal. The user can receive this information on their smartphone or tablet. The input is recommendation information, and the output is visualized recommendation information.

[0385] Step 8:

[0386] Users utilize a facility or store and then provide feedback. The terminal sends this feedback to a server. The input is the user's feedback, and the output is the analyzed feedback information.

[0387] Step 9:

[0388] The server stores the feedback it receives in a database and uses it for retraining. This allows the AI ​​model to continuously learn and improve its recommendation accuracy. The input is the analyzed feedback information, and the output is the updated AI model.

[0389] The above outlines the specific processing steps of the system that implements the application example.

[0390] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0391] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0392] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0393] [Second Embodiment]

[0394] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0395] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0396] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0397] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0398] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0399] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0400] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0401] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0402] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0403] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0404] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0405] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0406] This invention is a system for collecting local information data from local governments and individual businesses and recommending the most suitable facilities and stores based on user requests. This system operates based on the interaction between a server, a terminal, and a user.

[0407] Collection and storage of local data

[0408] The server receives local information data sent from local governments and individual businesses. The received data is appropriately analyzed and classified into categories such as "facilities," "events," and "restaurants." This classified data is stored in a database. This data forms the basis for generating recommendation information in response to user requests.

[0409] Data preprocessing and AI model training

[0410] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is formatted and converted into a format suitable for machine learning algorithms. This formatted data is used to train the AI ​​model. The newly trained AI model is deployed to the production environment in a timely manner. This enables the provision of highly accurate recommendation information to users.

[0411] Processing user requests

[0412] Users request information about facilities and stores through LINE or other applications. The device sends this request to a server. The server analyzes the received request and searches its database for relevant information. Based on the retrieved information, an AI model generates the most suitable recommendations for the user. These recommendations are then delivered to the user through their device.

[0413] As a concrete example, a user sends a request via the LINE app saying, "Tell me your recommended ramen restaurant." The device sends this request to the server. The server searches its database for information on relevant ramen restaurants and uses an AI model to make the best recommendation. For example, the user might be provided with a recommendation such as, "The ramen restaurant 'Mensho' near the station has high ratings."

[0414] Gathering feedback and relearning

[0415] After a user uses a recommended facility or store, they provide feedback. The device sends this feedback to the server. The server stores the feedback data in a database and uses it for retraining. This new feedback data is also integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[0416] As a concrete example, a user sends feedback saying "It was very delicious" about a ramen restaurant they visited. This feedback is sent to the server and stored in the database. The server uses this feedback data to retrain the AI ​​model and incorporate it into future recommendations.

[0417] Thus, the present invention is a system that efficiently collects local information from local governments and individual shops and provides users with highly accurate recommendation information, thereby stimulating local consumer activity and improving the likelihood of people staying in the area.

[0418] The following describes the processing flow.

[0419] Collection and storage of local data

[0420] Step 1:

[0421] The server receives HTTP POST requests from local governments and individual businesses.

[0422] Step 2:

[0423] The server parses the received request body and extracts regional information data.

[0424] Step 3:

[0425] The server categorizes the extracted data into categories such as "facilities," "events," and "restaurants."

[0426] Step 4:

[0427] The server stores the classified data in a database.

[0428] Data preprocessing and AI model training

[0429] Step 1:

[0430] The server periodically checks the raw data in the database, corrects incomplete data, and removes noisy data.

[0431] Step 2:

[0432] The server formats the cleansed data into a format suitable for machine learning algorithms.

[0433] Step 3:

[0434] The server uses the formatted data to train the AI ​​model.

[0435] Step 4:

[0436] The server evaluates the model's performance and adjusts parameters as needed.

[0437] Step 5:

[0438] The server uploads the newly trained model to the production deployment environment.

[0439] Processing user requests

[0440] Step 1:

[0441] Users request information about facilities and stores through LINE or other apps.

[0442] Step 2:

[0443] The terminal sends the user's request to the server as an HTTP request.

[0444] Step 3:

[0445] The server parses the received request and understands its contents.

[0446] Step 4:

[0447] The server searches the database for information related to the request.

[0448] Step 5:

[0449] The server uses an AI model to generate optimal recommendations based on the search results.

[0450] Step 6:

[0451] The server formats the generated recommendation information and sends it to the terminal.

[0452] Step 7:

[0453] The device displays recommendation information on the user interface.

[0454] Gathering feedback and relearning

[0455] Step 1:

[0456] Users provide feedback after using the recommended facilities or stores.

[0457] Step 2:

[0458] The device sends feedback to the server.

[0459] Step 3:

[0460] The server saves the feedback to the database.

[0461] Step 4:

[0462] The server integrates the new feedback data with the existing data.

[0463] Step 5:

[0464] The server retrains the AI ​​model using the integrated data.

[0465] In this way, the system achieves highly accurate information delivery through efficient processes of collecting, storing, analyzing, learning from, recommending, and providing feedback on local information.

[0466] (Example 1)

[0467] Next, we will describe Example 1. 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."

[0468] Conventional local information recommendation systems often suffered from incomplete data collection or reduced recommendation accuracy due to the inclusion of noisy data. Furthermore, responses to user requests were sometimes not in real time, compromising usability. In addition, insufficient retraining using feedback made continuous improvement of recommendation accuracy difficult. This invention aims to effectively solve these problems and improve the recommendation accuracy for local facilities and stores.

[0469] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0470] In this invention, the server includes means for receiving local information data from local governments and individual stores; means for analyzing the received local information data, classifying it into categories such as "facilities," "events," and "restaurants," and storing it in a database; means for periodically cleansing the stored local information data to correct defects and remove noise data; means for training an artificial intelligence model using a machine learning algorithm with the cleansed data; means for deploying the trained artificial intelligence model; means for analyzing user requests using natural language processing technology; means for searching for relevant information from the database based on the analyzed requests and generating recommendation information using the artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; and means for receiving user feedback, storing the received feedback in a database, and using it for retraining. This improves the accuracy of local information data collection and cleansing, and enables real-time responses. Furthermore, the recommendation accuracy continuously improves by incorporating feedback.

[0471] "Local information data" refers to various types of information provided by local governments and individual businesses, such as facility names, addresses, contact information, and event information.

[0472] "Analysis" refers to the process of appropriately assigning and classifying received data into each category.

[0473] A "category" refers to a specific group used to classify information such as "facilities," "events," and "restaurants."

[0474] A "database" refers to a data management system for efficiently storing, searching, and updating information.

[0475] "Cleansing" refers to the process of correcting data defects and removing noisy data.

[0476] A "machine learning algorithm" refers to a mathematical model that uses large amounts of data to detect patterns and makes predictions and classifications based on new data.

[0477] An "artificial intelligence model" refers to a system that uses machine learning algorithms to analyze data, make predictions, and generate recommendation information.

[0478] "Deployment" refers to introducing a trained artificial intelligence model into a production environment and putting it into operation.

[0479] "Natural language processing technology" refers to computer science techniques used to analyze human language and understand its meaning.

[0480] "Recommendation information" refers to information about the most suitable facilities and stores, generated based on user requests.

[0481] "Feedback" refers to information such as ratings and comments that users provide after using a recommended facility or store.

[0482] "Retraining" refers to the process of retraining an artificial intelligence model using newly collected data and feedback data to improve the model's accuracy and performance.

[0483] This invention is a system that collects local information data from local governments and individual businesses and recommends the most suitable facilities and businesses based on user requests. This system operates based on the interaction of a server, terminals, and users.

[0484] Collection and storage of local data

[0485] The server receives local information data sent from local governments and individual businesses. This data includes facility names, addresses, contact information, and event information. The received data is analyzed by the server and classified into categories such as "facilities," "events," and "restaurants." The classified data is stored in a database (e.g., MySQL or PostgreSQL). This data forms the basis for generating recommendation information in response to user requests.

[0486] Data preprocessing and AI model training

[0487] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is converted into a format suitable for machine learning algorithms. For example, the Google Maps API is used to verify the accuracy of addresses to compensate for data defects. Using the formatted data, the server trains an AI model (e.g., using TensorFlow or PyTorch). After training is complete, the newly trained AI model is deployed to the production environment.

[0488] Processing user requests

[0489] Users request information about specific facilities or stores via LINE or a dedicated app. For example, they might send a request like, "Tell me your recommended ramen restaurant." The device sends this request to the server in real time. The server analyzes the received request using natural language processing technology (e.g., Google NLP API or SpaCy). It then searches the database for relevant information, and an AI model generates the most suitable recommendation information.

[0490] Providing recommendation information

[0491] The server sends the generated recommendation information to the device. For example, the user is provided with a recommendation such as, "The ramen shop in front of the station has high ratings." The device then displays this information to the user. In the case of LINE, it is displayed as a message, and in the case of a dedicated app, it is displayed as an in-app notification or on the screen.

[0492] Gathering feedback and relearning

[0493] Users provide feedback after using recommended facilities or stores. For example, they might send a simple rating such as "It was delicious." The device sends this feedback to the server. The server stores the feedback data in a database and uses it for training the next AI model. The AI ​​model, retrained with the new feedback data, generates even more accurate recommendations.

[0494] Specific example

[0495] A user sends a request via the LINE app saying, "Tell me your recommended ramen restaurant." The device sends this request to the server. The server searches its database for information on relevant ramen restaurants and uses an AI model to make the best recommendation. For example, the user is provided with a recommendation such as, "The ramen restaurant '○○' near the station has high ratings."

[0496] By presenting specific situations in this way, the generative AI model becomes more likely to generate specific and appropriate flows.

[0497] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0498] Step 1: Collection and reception of local information data

[0499] The server receives local information data sent from local governments and individual businesses via APIs and database connections. Specifically, the server uses API crawling and scraping techniques to retrieve data such as facility names, addresses, contact information, and event information. The input is data from local governments and individual businesses, and the output is raw local information data stored in the server's internal database.

[0500] Step 2: Analysis and classification of regional information data

[0501] The server analyzes the received data using natural language processing techniques (e.g., Google NLP API or SpaCy). Specifically, it performs text analysis and keyword matching to classify the data into categories such as "facilities," "events," and "restaurants." The input is the raw data received in step 1, and the output is the data categorized.

[0502] Step 3: Save data

[0503] The server stores the classified data in a relational database (e.g., MySQL or PostgreSQL). Specifically, the server uses SQL queries to store the data in the appropriate tables within the database. The input is the data classified in step 2, and the output is the structured data stored in the database.

[0504] Step 4: Data Cleansing

[0505] The server periodically checks the data in the database, corrects incomplete data, and removes noisy data. Specifically, it uses the Google Maps API to fill in missing address information and standardizes data with inconsistent formats. The input is existing data in the database, and the output is cleansed and formatted data.

[0506] Step 5: Training the AI ​​model

[0507] The server uses the cleansed data to train an AI model using machine learning algorithms (e.g., TensorFlow or PyTorch). Specifically, it processes large amounts of data on a GPU (e.g., NVIDIA Tesla V100) and learns patterns to generate the expected output. The input is the cleansed data, and the output is the trained AI model.

[0508] Step 6: Deploying the Artificial Intelligence Model

[0509] The server deploys the trained AI model to the production environment. Specifically, it uploads the model to the server and makes it accessible via API. The input is the trained AI model, and the output is the publicly available AI model.

[0510] Step 7: Receiving and parsing user requests

[0511] Users request information about specific facilities or stores via LINE or a dedicated application. The device sends this request to the server in real time. The server analyzes the received request using natural language processing technology and converts it into structured data. The input is the user request, and the output is the analyzed request data.

[0512] Step 8: Generating and searching for recommendation information

[0513] The server searches the database for relevant information based on the parsed request and generates optimal recommendations using an AI model. Specifically, the server uses collaborative filtering and content-based filtering to recommend the most suitable facilities and stores to the user. The input is the parsed request data and relevant data in the database, and the output is the recommendations.

[0514] Step 9: Submitting and displaying recommendation information

[0515] The server sends the generated recommendation information to the device. The device then displays this information to the user. Specifically, in the case of LINE, it is provided to the user as a message, and in the case of a dedicated app, it is provided as an in-app notification or screen display. The input is the recommendation information, and the output is the recommendation content provided to the user.

[0516] Step 10: Gathering Feedback and Retraining

[0517] Users provide feedback after using recommended facilities or stores. Specifically, users send ratings and comments via a dedicated app or LINE. The device sends this feedback to a server. The server stores the feedback data in a database and retrains the AI ​​model by incorporating the new data. The input is the user's feedback, and the output is the retrained AI model.

[0518] (Application Example 1)

[0519] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0520] Traditional facility and store recommendation systems suffer from insufficient collection and analysis of local information, making it difficult to respond quickly and appropriately to user requests. Furthermore, the inability to provide personalized recommendations based on user preferences and usage history makes improving user satisfaction a challenge. Additionally, the inability to offer value-added services such as real-time event notifications makes it difficult to maintain user interest.

[0521] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0522] In this invention, the server includes means for receiving local information data from local governments and individual stores; means for classifying the received local information data and storing it in a database; means for training an artificial intelligence model to generate facility and store recommendations based on user requests using the stored local information data; means for receiving requests from users; means for searching the database for relevant information based on user requests and generating recommendations using the artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; means for receiving feedback from users, storing the received feedback in a database and using it for retraining; means for processing user requests via an application installed on a smartphone; and means for providing information on recommended stores and real-time event notifications via push notifications. This enables personalized recommendations of facilities and stores tailored to the user's interests and preferences, and improves the user experience through real-time information provision and feedback reflection.

[0523] A "local government" is a public institution that carries out administrative duties in a specific region and is a source of regional information data.

[0524] A "personal store" is a small-scale commercial facility or service provider owned or operated by an individual, and is a source of local information data.

[0525] "Local information data" refers to information about facilities, shops, events, and services located within a specific area, and is provided by local governments and individual businesses.

[0526] A "database" is a digital system for organizing and storing regional information data in a searchable and accessible format.

[0527] An "artificial intelligence model" is an algorithm or computational model that learns from a large amount of data and performs a specific task; in this context, it refers to a model used in recommendation systems.

[0528] A "user terminal" refers to a device used by a user, such as a computer, smartphone, or tablet, which is used to receive recommendation information and send requests.

[0529] "Feedback" refers to opinions and impressions, such as ratings and comments, provided by users, and is information used for system improvement and retraining.

[0530] "Retraining" is the process of updating an existing artificial intelligence model using newly collected data and feedback to improve its performance and accuracy.

[0531] A "smartphone" is a portable computer with mobile communication and internet connectivity capabilities, and is a device that allows for the installation and use of applications.

[0532] "Push notifications" are notification messages automatically sent from a server to a user's device, providing a means of delivering important information and updates in real time.

[0533] A "user interface" is an interface through which a user interacts with a system, providing information using messaging applications and push notification functions.

[0534] To implement this invention, interaction between a server, a terminal, and a user is necessary. A detailed embodiment is shown below.

[0535] Collection and storage of local data

[0536] First, the server receives local information data from local governments and individual businesses. This data is categorized into categories such as "facilities," "events," and "restaurants." The categorized data is stored in a database such as Firebase. This data forms the basis for generating recommendation information in response to user requests.

[0537] Data preprocessing and AI model training

[0538] Next, the server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is then formatted and converted into a format suitable for machine learning algorithms. Using this formatted data, an AI model is trained using Python libraries such as Scikit-learn. The newly trained AI model is deployed to the production environment as needed. This enables the provision of highly accurate recommendation information to users.

[0539] Processing user requests

[0540] Users can request information about facilities and shops through applications or messaging apps installed on their smartphones. In this case, if a user sends a request such as "Tell me a good ramen shop," the device sends this request to the server. The server analyzes the received request and searches its database for relevant information. Based on the retrieved information, an AI model generates the most suitable recommendations for the user. The generated recommendations are then sent as push notifications to the user's smartphone, for example, "The ramen shop in front of the station has high ratings."

[0541] Gathering feedback and relearning

[0542] After a user visits a recommended facility or store, the terminal sends feedback to the server. For example, a comment such as "It was delicious" might be sent. The server stores this feedback data in a database and uses it for retraining. This new feedback data is also integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[0543] Specific example

[0544] As a concrete example, when a user sends a request using a smartphone application saying, "Tell me some recommended cafes," the server searches for relevant cafes in real time, and the AI ​​model can recommend cafes with high ratings. The user is notified of the recommendation, such as, "The cafe near the station has high ratings." After the user visits the cafe, they can send feedback such as, "It was very comfortable," and this information will be used to train the AI ​​model for the next time.

[0545] This system will allow users to receive recommendations for facilities and shops that best suit their interests and preferences, and is expected to further stimulate local consumer activity.

[0546] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0547] Step 1:

[0548] The server receives local information data from local governments and individual businesses and stores it in a database. Specifically, it retrieves data via CSV files or APIs and stores it in the database as structured data. The input is local information data, and the output is organized and categorized data.

[0549] Step 2:

[0550] The server periodically checks the regional information data in the database, corrects incomplete data, and removes noisy data. For example, it imputes missing values ​​and removes inappropriate values. This process results in cleansed data. The input is the regional information data in the database, and the output is the cleansed data.

[0551] Step 3:

[0552] The server uses the cleansed data to transform it into a format suitable for machine learning algorithms and trains an AI model. It uses Python's Scikit-learn to vectorize the data and train the model. The input is the cleansed data, and the output is the trained AI model.

[0553] Step 4:

[0554] Users request information about facilities and stores through a smartphone application. For example, they might type "Tell me your recommended ramen restaurant" as text. The input is the user's request, and the output is the request data.

[0555] Step 5:

[0556] The terminal sends the user's request to the server. The server parses the request and searches the database for relevant information. For example, it might perform a search based on the keyword "ramen shop." The input is the request data, and the output is the relevant information.

[0557] Step 6:

[0558] The server generates optimal recommendations using an AI model based on the searched information. It uses a trained model to select highly-rated stores. The input is relevant information, and the output is recommendations.

[0559] Step 7:

[0560] The server sends the generated recommendation information to the user's terminal. For example, it might send a push notification saying, "The ramen shop in front of the station has high ratings." The input is the recommendation information, and the output is the notification sent to the user's terminal.

[0561] Step 8:

[0562] Users visit recommended facilities or shops and then provide feedback. They submit comments such as "It was delicious." The input is the user's feedback, and the output is feedback data.

[0563] Step 9:

[0564] The device sends user feedback to the server. The server stores the feedback data in a database and uses it for retraining. The collected feedback data is also integrated with existing data and used to retrain the AI ​​model. The input is the feedback data, and the output is the updated AI model.

[0565] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0566] This invention is a system for collecting local information data from local governments and individual businesses, recognizing user emotions using an emotion engine based on user requests and feedback, and utilizing that information to recommend the most suitable facilities and businesses. This system operates based on the interaction between a server, a terminal, and a user.

[0567] Collection and storage of local data

[0568] The server receives local information data sent from local governments and individual businesses. The received data is appropriately analyzed, categorized into categories such as "facilities," "events," and "restaurants," and stored in a database. This data forms the basis for generating recommendation information in response to user requests.

[0569] Data preprocessing and AI model training

[0570] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is formatted and converted into a format suitable for machine learning algorithms. This formatted data is used to train the AI ​​model. The newly trained AI model is deployed to the production environment in a timely manner. This enables the provision of highly accurate recommendation information to users.

[0571] Processing user requests and sentiment analysis

[0572] Users request information about facilities and stores through LINE or other applications. The device sends this request to the server. The server analyzes the received request and understands its content. Furthermore, it can use an emotion engine to extract emotional information from the user's request. For example, if a user requests "I'm hungry and very irritated," the emotion engine recognizes "irritated."

[0573] Generation and provision of recommendation information

[0574] The server searches the database for information related to the request. Furthermore, using the retrieved sentiment information, an AI model generates recommendations best suited to the user. For example, a user who is "frustrated" might be recommended a restaurant that can provide service quickly. This recommendation information is then delivered to the user via their device.

[0575] As a concrete example, a user requests via the LINE app, "Tell me about a good ramen restaurant quickly." The device sends this request to the server. The server analyzes the request and uses an emotion engine to obtain "urgent" emotion information. The server searches its database for ramen restaurants that can respond quickly and uses an AI model to recommend, "The ramen restaurant 'Mensho' near the station is highly rated for its quick service." The device then provides this information to the user.

[0576] Gathering feedback and relearning

[0577] After a user visits a recommended facility or store, they provide feedback. The device sends this feedback to a server. The server stores the feedback data in a database and uses it for retraining. The sentiment information included in the provided feedback is also analyzed and stored in the database. This new feedback data is integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[0578] As a concrete example, a user might submit feedback about a ramen restaurant they visited, stating, "It was very delicious, but the service was a little slow." This feedback is sent to the server and stored in the database. The server uses this feedback data and the sentiment information analyzed by the sentiment engine to retrain the AI ​​model and incorporate it into future recommendations.

[0579] Thus, the present invention is a system that efficiently collects local information from local governments and individual shops, analyzes users' emotions using an emotion engine, and provides users with highly accurate recommendation information, thereby stimulating local consumer activity and improving their dwell time.

[0580] The following describes the processing flow.

[0581] Collection and storage of local data

[0582] Step 1:

[0583] The server receives HTTP POST requests from local governments and individual businesses.

[0584] Step 2:

[0585] The server parses the received request body and extracts regional information data.

[0586] Step 3:

[0587] The server categorizes the extracted data into categories such as "facilities," "events," and "restaurants."

[0588] Step 4:

[0589] The server stores the classified data in a database.

[0590] Data preprocessing and AI model training

[0591] Step 1:

[0592] The server periodically checks the raw data in the database, corrects incomplete data, and removes noisy data.

[0593] Step 2:

[0594] The server formats the cleansed data into a format suitable for machine learning algorithms.

[0595] Step 3:

[0596] The server uses the formatted data to train the AI ​​model.

[0597] Step 4:

[0598] The server evaluates the model's performance and adjusts parameters as needed.

[0599] Step 5:

[0600] The server uploads the newly trained model to the production deployment environment.

[0601] Processing user requests and sentiment analysis

[0602] Step 1:

[0603] Users request information about facilities and stores through LINE or other apps.

[0604] Step 2:

[0605] The terminal sends the user's request to the server as an HTTP request.

[0606] Step 3:

[0607] The server analyzes the received request to understand its content and the user's intent.

[0608] Step 4:

[0609] The server uses an emotion engine to extract emotional information from user requests.

[0610] Step 5:

[0611] The server searches the database for information related to the request.

[0612] Step 6:

[0613] The server uses an AI model to generate optimal recommendations based on search results and sentiment information.

[0614] Step 7:

[0615] The server formats the generated recommendation information and sends it to the terminal.

[0616] Step 8:

[0617] The device displays recommendation information on the user interface.

[0618] Specific example

[0619] Step 1:

[0620] The user sends a request in the LINE app saying "Please tell me a delicious ramen shop quickly."

[0621] Step 2:

[0622] The terminal sends this request to the server as an HTTP request.

[0623] Step 3:

[0624] The server analyzes the request and understands that the user is "in a hurry."

[0625] Step 4:

[0626] The server uses the emotion engine to extract the "in a hurry" emotion from the text of the request.

[0627] Step 5:

[0628] The server quickly searches the database for information on ramen shops that can provide services.

[0629] Step 6:

[0630] The server generates optimal recommendation information using an AI model based on the search results and emotion information.

[0631] Step 7:

[0632] The server formalizes the recommendation information "The ramen shop 'Noodle Master' in front of the station is evaluated for quick service" and sends it to the terminal.

[0633] Step 8:

[0634] The terminal displays this information to the user in the LINE app.

[0635] Collection of feedback and re - learning

[0636] Step 1:

[0637] Users provide feedback after using the recommended facilities or stores.

[0638] Step 2:

[0639] The device sends feedback to the server.

[0640] Step 3:

[0641] The server analyzes the feedback and extracts emotional information.

[0642] Step 4:

[0643] The server stores feedback data and emotional information in a database.

[0644] Step 5:

[0645] The server retrains the AI ​​model using the stored feedback data.

[0646] In this way, the system achieves highly accurate information provision and recommendations tailored to user sentiment through an efficient process of collecting, storing, analyzing, learning, recommending, and providing feedback on local information.

[0647] (Example 2)

[0648] Next, we will describe Example 2. 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".

[0649] Current regional information recommendation systems have a problem in that they struggle to make recommendations that take into account user emotions and urgency. Furthermore, the presence of data incompleteness and noise often leads to decreased recommendation accuracy. Additionally, there is a lack of effective means to utilize user feedback for retraining, which hinders the continuous improvement of system performance.

[0650] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0651] In this invention, the server includes means for receiving local information data from local governments and individual stores; means for classifying the received local information data and storing it in a database; means for checking the integrity of the stored local information data and correcting and deleting incomplete or noisy data; means for converting the formatted data into a format suitable for machine learning algorithms and training an artificial intelligence model; means for receiving requests from users; means for analyzing the content of requests and obtaining sentiment information using a sentiment engine; means for searching for relevant information from the database and generating recommendation information using an artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; means for receiving feedback from users, storing the received feedback in a database and using it for retraining; and means for optimizing the recommendation information based on the generated sentiment information. This makes it possible to provide highly accurate recommendation information based on the user's sentiment and urgency.

[0652] A "local government" refers to a public institution responsible for managing and operating a specific region.

[0653] "Individual store" refers to a commercial or service facility operated by an individual or a small business.

[0654] "Local information data" refers to data that includes information about facilities, events, restaurants, etc., related to a specific region.

[0655] A "database" refers to a system for systematically storing and managing collected information.

[0656] "Incomplete data" refers to data that is missing necessary information or contains errors.

[0657] "Noise data" refers to unwanted data that hinders analysis and learning.

[0658] A "machine learning algorithm" refers to a method or model for automatically learning patterns and rules from data.

[0659] An "artificial intelligence model" refers to a program that has been trained using machine learning algorithms and possesses the ability to perform specific tasks.

[0660] An "emotion engine" refers to a technology that uses natural language processing and text mining to recognize and analyze emotions from text.

[0661] A "user terminal" refers to an electronic device (e.g., smartphone, tablet, personal computer) used by a user to input and receive information.

[0662] A "request" refers to a request made by a user to a system for the provision of information.

[0663] "Feedback" refers to the evaluations and opinions that users give regarding the information and services provided.

[0664] "Retraining" refers to the process of retraining an artificial intelligence model using new data and feedback.

[0665] "Optimization" refers to adjusting the parameters and processes of a system to their optimal state in order to achieve a specific objective.

[0666] This invention relates to a system that collects local information data from local governments and individual businesses and generates recommendation information based on user requests and feedback. This system operates based on the interaction between a server, a terminal, and a user.

[0667] Collection and storage of local data

[0668] The server receives local information data sent from local governments and individual businesses. This information is retrieved through interfaces such as APIs. Examples include event information from local governments and business hours from individual businesses. This data is categorized into categories such as "facilities," "events," and "restaurants," and stored in a database. Relational databases such as MySQL and PostgreSQL can be used as the database. This data later forms the basis for generating recommendation information in response to user requests.

[0669] Data preprocessing and AI model training

[0670] The server periodically checks the received regional data, correcting and removing incomplete and noisy data. Specifically, it fills in missing information and removes outliers. Data cleansing scripts using Python or Pandas are used. The formatted data is converted into a format suitable for machine learning algorithms. For example, categorical data is encoded into numerical data. This is then used to train the AI ​​model. TensorFlow and PyTorch are used as machine learning frameworks. The newly trained AI model is deployed to the production environment as needed.

[0671] Processing user requests and sentiment analysis

[0672] Users request information about facilities and stores through LINE or a dedicated application. For example, they might send a message like, "Tell me about a good ramen restaurant quickly." The device sends this request to the server. The server analyzes the received request and uses natural language processing technology (e.g., spaCy or NLTK) to understand its content. Furthermore, it uses an emotion engine (e.g., Google Cloud Natural Language API) to extract emotional information from the user's request. For example, it recognizes "irritated" from a request like, "I'm hungry and very irritated."

[0673] Generation and provision of recommendation information

[0674] The server searches the database for information related to the request, and the AI ​​model uses the retrieved sentiment information to generate the most suitable recommendations for the user. For example, for a user who is "frustrated," it recommends restaurants that can provide quick service. This recommendation information is provided to the user through their device. As a specific example, if a user requests "Tell me a good ramen shop quickly" using the LINE app, the device sends the request to the server, and the server analyzes the request and sentiment information. It searches the database for ramen shops that can respond quickly and uses the AI ​​model to recommend "Ramen shop 'XX' near the station is highly rated for its quick service." This information is provided to the user through their device.

[0675] Gathering feedback and relearning

[0676] Users provide feedback after using recommended facilities or stores. For example, they might send feedback via the LINE app, such as, "It was delicious, but the service was a little slow." The device sends this feedback to a server. The server stores the received feedback data in a database and uses it for retraining. The collected feedback data undergoes cleansing and formatting steps, is integrated with existing data, and then used to retrain the AI ​​model. This ensures that the system can always provide highly accurate recommendation information.

[0677] Examples of prompt statements

[0678] The following are specific examples of input prompts for a generative AI model.

[0679] 1. "Can you recommend a good cafe nearby?"

[0680] 2. "I'm looking for a restaurant that's safe and comfortable for families with children."

[0681] 3. "What are some recommended events to go to with friends?"

[0682] 4. "Please recommend a good place for a date."

[0683] By utilizing these prompts, the generative AI model can provide optimal recommendation information based on user requests.

[0684] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0685] Step 1: Collecting local data

[0686] The server receives local information data from local governments and individual businesses. The specific operation is as follows: First, input processing is performed to acquire data through an interface such as an API. Next, the received data is temporarily stored in a buffer. Then, it is classified into categories such as "facilities," "events," and "restaurants," and finally stored in the database. In this step, the input is the received local information data, and the output is the classified and stored data.

[0687] Step 2: Data preprocessing

[0688] The server periodically checks the received regional information data, correcting and removing incomplete and noisy data. Specifically, it performs data integrity checks, supplements missing information, and removes outliers. Next, it formats this data into a format suitable for machine learning algorithms. For example, it encodes categorical data into numerical data. In this step, the input is classified data, and the output is formatted data.

[0689] Step 3: Training the AI ​​model

[0690] The server trains an AI model using pre-processed data. The specific operation is as follows: First, the new data is split into a training set and a test set. Next, the AI ​​model is trained using a machine learning framework such as TensorFlow or PyTorch. The trained model is finally deployed to the production environment. The input in this step is a formatted dataset, and the output is the trained AI model.

[0691] Step 4: Processing User Requests

[0692] Users send requests via LINE or a dedicated application. A concrete example of such a request might be, "Tell me a good ramen restaurant quickly." The device sends this request to the server. The server uses natural language processing techniques to analyze the received request and understand its content. In this step, the input is the user request, and the output is the analyzed request content.

[0693] Step 5: Emotion Analysis

[0694] The server analyzes the received request using an emotion engine to obtain the user's emotion information. Specifically, it uses a natural language processing library to extract keywords and emotional expressions from the request and obtains the emotion information through the emotion engine. For example, from the request "Tell me a good ramen restaurant quickly," the emotion information "I'm in a hurry" is extracted. In this step, the input is the analyzed request content, and the output is the obtained emotion information.

[0695] Step 6: Generating recommendation information

[0696] The server searches the database for relevant information and generates optimal recommendations based on the retrieved sentiment information and user requests. Specifically, it uses an AI model to evaluate the request content and sentiment data to generate optimal recommendations. For example, it might recommend stores that can provide quick service to a user who is "in a hurry." The inputs in this step are sentiment information and user requests, and the output is the generated recommendations.

[0697] Step 7: Providing Recommendation Information

[0698] The server provides the generated recommendation information to the user via the terminal. Specifically, the recommendation information is sent to the terminal via an API, and the terminal displays that information to the user. For example, a recommendation such as "The ramen shop 'XX' in front of the station is highly rated for its fast service" is notified to the LINE app. In this step, the input is the generated recommendation information, and the output is the notification to the user.

[0699] Step 8: Gathering Feedback

[0700] Users provide feedback after using a facility or store. For example, they might say, "The food was delicious, but the service was a little slow." The device sends this feedback to the server, which then stores it in a database. In this step, the input is the user feedback, and the output is the stored feedback data.

[0701] Step 9: Retraining the AI ​​model

[0702] The server retrains the AI ​​model using the collected feedback data. Specifically, it cleanses and formats the new feedback data and integrates it with the existing dataset. Then, it retrains the AI ​​model and deploys it to the production environment. In this step, the input is the saved feedback data, and the output is the retrained AI model.

[0703] (Application Example 2)

[0704] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0705] Traditional local information recommendation systems have a problem of not providing sufficient user satisfaction because they simply provide information without considering the user's emotional state. Furthermore, online virtual shopping lacks personalized recommendations based on user emotions and feedback. This results in a uniform shopping experience for all users, making it difficult to recommend optimal products and services that meet individual needs.

[0706] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for receiving local information data from local governments and individual stores; means for classifying the received local information data and storing it in a database; means for training an artificial intelligence model for generating facility and store recommendations based on user requests using the stored local information data; means for searching for relevant information from the database based on user requests and sentiment information and generating recommendations using the artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; means for receiving user feedback, storing the received feedback in a database and using it for retraining; and means for recommending products and services in the virtual store based on the stored local information data and user sentiment information. This makes it possible to provide detailed information recommendations based on user sentiment and feedback, thereby improving user satisfaction and the shopping experience.

[0707] A "local government" is an administrative body that governs a specific region and is responsible for providing public services and promoting the development of the local community.

[0708] A "private store" is a commercial facility owned and operated by an individual, primarily providing goods or services.

[0709] "Local information data" refers to data provided by local governments or individual businesses that includes various types of information about a particular region.

[0710] A "database" is a system for efficiently storing, searching, and managing data.

[0711] "Emotional information" refers to data that indicates a user's emotional state, and is primarily analyzed by an emotion engine.

[0712] A "user terminal" is a device used by a user to interact with the system, and includes smartphones, tablets, and other similar devices.

[0713] An "artificial intelligence model" is an algorithm or system that automatically performs a specific task by analyzing and learning from data.

[0714] "Feedback" refers to data, including user opinions and impressions after using a system, which is used to improve the system.

[0715] A "virtual store" is a commercial facility in an online environment that provides goods and services via the internet.

[0716] "Recommendation information" refers to information about the most suitable facilities and stores, generated based on user requests and sentiments.

[0717] "Retraining" is the process of retraining an existing artificial intelligence model using new data and feedback.

[0718] This invention is a system that collects local information data and generates and provides recommendation information based on user requests and sentiment information. Specific embodiments of this invention are described below.

[0719] 1. Data collection and storage

[0720] The server receives local information data from local governments and individual businesses. This data includes a wide range of information such as facilities, events, and restaurants. The received data is appropriately analyzed, categorized, and then stored in a database.

[0721] 2. Data preprocessing and AI model training

[0722] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is formatted to a format suitable for machine learning algorithms, and this is used to train AI models. The newly trained AI models are deployed to the production environment, enabling them to provide users with highly accurate recommendations.

[0723] 3. Processing User Requests and Sentiment Analysis

[0724] Users request information through the virtual store's shopping app or messaging application. The device sends this request to the server. The server analyzes the received request and uses an emotion engine to obtain the user's emotional information. For example, if a user requests, "I'm feeling stressed today, so I'm looking for a relaxing cafe. Do you have any recommendations?", the emotion engine recognizes "stress."

[0725] 4. Generation and provision of recommendation information

[0726] The server searches the database for information related to the request, and using the retrieved sentiment information, the AI ​​model generates recommendations best suited to the user. For example, a user feeling "stressed" would be recommended relaxing cafes or shops. This recommendation information is then sent to the user's device.

[0727] 5. Gathering feedback and relearning

[0728] After a user uses a recommended facility or store, they provide feedback. The device sends this feedback to a server. The server stores the feedback data in a database and uses it for retraining. The sentiment information included in the provided feedback is also analyzed and stored in the database. This new feedback data is integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[0729] Hardware and software to be used

[0730] Hardware: Servers, smartphones, tablets

[0731] Software: Web frameworks (e.g., Flask, Django), machine learning libraries (e.g., TensorFlow), sentiment analysis engines (e.g., Emotion-Recognition-API)

[0732] Specific example

[0733] When a user requests, "I'm feeling stressed today, so I'm looking for a relaxing cafe. Do you have any recommendations?", the emotion engine recognizes the user's "stress" and recommends a relaxing establishment. Similarly, if a user requests, "Tell me about a good ramen shop quickly," the system can recommend a ramen shop that can provide quick service based on the emotional information that the user is "in a hurry."

[0734] Example of a prompt:

[0735] "I'm feeling stressed today, so I'm looking for a relaxing cafe. Do you have any recommendations?"

[0736] "Please tell me a good ramen shop quickly. Thank you."

[0737] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0738] Step 1:

[0739] The server receives local information data from local governments and individual businesses. This data is often provided in a structured format such as JSON. The received data is saved to a database in real time. The input is local information data sent from local governments and individual businesses, and the output is the saved data.

[0740] Step 2:

[0741] The server classifies the data in the database and stores it by category. For example, it might classify data into categories such as "restaurants," "events," and "facilities." This classification is performed using a data analysis algorithm. The input is the stored regional information data, and the output is the classified data.

[0742] Step 3:

[0743] The server uses the cleansed data to train an artificial intelligence model. Data shaping processes convert the data into a format suitable for machine learning algorithms, and this is then input into the AI ​​model. This dataset is used to train the model and build a highly accurate recommendation engine. The input is the shaped data, and the output is the trained AI model.

[0744] Step 4:

[0745] The user sends a request via a smartphone or tablet. The request content is parsed using natural language processing. The device sends this request to the server, and the prompt text includes the user's sentiment. The input is the user request, and the output is the parsing result.

[0746] Step 5:

[0747] The server analyzes the received request and uses an emotion engine to obtain the user's emotional information. For example, the emotion engine analyzes emotional information such as "I want to relax," "I'm in a hurry," or "I'm stressed." The input is the user's request, and the output is the user's emotional information.

[0748] Step 6:

[0749] The server searches the database for relevant information and uses an AI model to generate optimal recommendations based on the retrieved sentiment information. For example, based on sentiment information such as "I want to relax," it might recommend quiet cafes or parks. The input is the user's sentiment information and request, and the output is the recommendations.

[0750] Step 7:

[0751] The server sends the generated recommendation information to the user's terminal. The user can receive this information on their smartphone or tablet. The input is recommendation information, and the output is visualized recommendation information.

[0752] Step 8:

[0753] Users utilize a facility or store and then provide feedback. The terminal sends this feedback to a server. The input is the user's feedback, and the output is the analyzed feedback information.

[0754] Step 9:

[0755] The server stores the feedback it receives in a database and uses it for retraining. This allows the AI ​​model to continuously learn and improve its recommendation accuracy. The input is the analyzed feedback information, and the output is the updated AI model.

[0756] The above outlines the specific processing steps of the system that implements the application example.

[0757] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0758] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0759] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0760] [Third Embodiment]

[0761] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0762] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0763] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0764] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0765] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0766] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0767] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0768] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0769] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0770] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0771] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0772] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0773] This invention is a system for collecting local information data from local governments and individual businesses and recommending the most suitable facilities and stores based on user requests. This system operates based on the interaction between a server, a terminal, and a user.

[0774] Collection and storage of local data

[0775] The server receives local information data sent from local governments and individual businesses. The received data is appropriately analyzed and classified into categories such as "facilities," "events," and "restaurants." This classified data is stored in a database. This data forms the basis for generating recommendation information in response to user requests.

[0776] Data preprocessing and AI model training

[0777] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is formatted and converted into a format suitable for machine learning algorithms. This formatted data is used to train the AI ​​model. The newly trained AI model is deployed to the production environment in a timely manner. This enables the provision of highly accurate recommendation information to users.

[0778] Processing user requests

[0779] Users request information about facilities and stores through LINE or other applications. The device sends this request to a server. The server analyzes the received request and searches its database for relevant information. Based on the retrieved information, an AI model generates the most suitable recommendations for the user. These recommendations are then delivered to the user through their device.

[0780] As a concrete example, a user sends a request via the LINE app saying, "Tell me your recommended ramen restaurant." The device sends this request to the server. The server searches its database for information on relevant ramen restaurants and uses an AI model to make the best recommendation. For example, the user might be provided with a recommendation such as, "The ramen restaurant 'Mensho' near the station has high ratings."

[0781] Gathering feedback and relearning

[0782] After a user uses a recommended facility or store, they provide feedback. The device sends this feedback to the server. The server stores the feedback data in a database and uses it for retraining. This new feedback data is also integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[0783] As a concrete example, a user sends feedback saying "It was very delicious" about a ramen restaurant they visited. This feedback is sent to the server and stored in the database. The server uses this feedback data to retrain the AI ​​model and incorporate it into future recommendations.

[0784] Thus, the present invention is a system that efficiently collects local information from local governments and individual shops and provides users with highly accurate recommendation information, thereby stimulating local consumer activity and improving the likelihood of people staying in the area.

[0785] The following describes the processing flow.

[0786] Collection and storage of local data

[0787] Step 1:

[0788] The server receives HTTP POST requests from local governments and individual businesses.

[0789] Step 2:

[0790] The server parses the received request body and extracts regional information data.

[0791] Step 3:

[0792] The server categorizes the extracted data into categories such as "facilities," "events," and "restaurants."

[0793] Step 4:

[0794] The server stores the classified data in a database.

[0795] Data preprocessing and AI model training

[0796] Step 1:

[0797] The server periodically checks the raw data in the database, corrects incomplete data, and removes noisy data.

[0798] Step 2:

[0799] The server formats the cleansed data into a format suitable for machine learning algorithms.

[0800] Step 3:

[0801] The server uses the formatted data to train the AI ​​model.

[0802] Step 4:

[0803] The server evaluates the model's performance and adjusts parameters as needed.

[0804] Step 5:

[0805] The server uploads the newly trained model to the production deployment environment.

[0806] Processing user requests

[0807] Step 1:

[0808] Users request information about facilities and stores through LINE or other apps.

[0809] Step 2:

[0810] The terminal sends the user's request to the server as an HTTP request.

[0811] Step 3:

[0812] The server parses the received request and understands its contents.

[0813] Step 4:

[0814] The server searches the database for information related to the request.

[0815] Step 5:

[0816] The server uses an AI model to generate optimal recommendations based on the search results.

[0817] Step 6:

[0818] The server formats the generated recommendation information and sends it to the terminal.

[0819] Step 7:

[0820] The device displays recommendation information on the user interface.

[0821] Gathering feedback and relearning

[0822] Step 1:

[0823] Users provide feedback after using the recommended facilities or stores.

[0824] Step 2:

[0825] The device sends feedback to the server.

[0826] Step 3:

[0827] The server saves the feedback to the database.

[0828] Step 4:

[0829] The server integrates the new feedback data with the existing data.

[0830] Step 5:

[0831] The server retrains the AI ​​model using the integrated data.

[0832] In this way, the system achieves highly accurate information delivery through efficient processes of collecting, storing, analyzing, learning from, recommending, and providing feedback on local information.

[0833] (Example 1)

[0834] Next, we will describe Example 1. 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."

[0835] Conventional local information recommendation systems often suffered from incomplete data collection or reduced recommendation accuracy due to the inclusion of noisy data. Furthermore, responses to user requests were sometimes not in real time, compromising usability. In addition, insufficient retraining using feedback made continuous improvement of recommendation accuracy difficult. This invention aims to effectively solve these problems and improve the recommendation accuracy for local facilities and stores.

[0836] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0837] In this invention, the server includes means for receiving local information data from local governments and individual stores; means for analyzing the received local information data, classifying it into categories such as "facilities," "events," and "restaurants," and storing it in a database; means for periodically cleansing the stored local information data to correct defects and remove noise data; means for training an artificial intelligence model using a machine learning algorithm with the cleansed data; means for deploying the trained artificial intelligence model; means for analyzing user requests using natural language processing technology; means for searching for relevant information from the database based on the analyzed requests and generating recommendation information using the artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; and means for receiving user feedback, storing the received feedback in a database, and using it for retraining. This improves the accuracy of local information data collection and cleansing, and enables real-time responses. Furthermore, the recommendation accuracy continuously improves by incorporating feedback.

[0838] "Local information data" refers to various types of information provided by local governments and individual businesses, such as facility names, addresses, contact information, and event information.

[0839] "Analysis" refers to the process of appropriately assigning and classifying received data into each category.

[0840] A "category" refers to a specific group used to classify information such as "facilities," "events," and "restaurants."

[0841] A "database" refers to a data management system for efficiently storing, searching, and updating information.

[0842] "Cleansing" refers to the process of correcting data defects and removing noisy data.

[0843] A "machine learning algorithm" refers to a mathematical model that uses large amounts of data to detect patterns and makes predictions and classifications based on new data.

[0844] An "artificial intelligence model" refers to a system that uses machine learning algorithms to analyze data, make predictions, and generate recommendation information.

[0845] "Deployment" refers to introducing a trained artificial intelligence model into a production environment and putting it into operation.

[0846] "Natural language processing technology" refers to computer science techniques used to analyze human language and understand its meaning.

[0847] "Recommendation information" refers to information about the most suitable facilities and stores, generated based on user requests.

[0848] "Feedback" refers to information such as ratings and comments that users provide after using a recommended facility or store.

[0849] "Retraining" refers to the process of retraining an artificial intelligence model using newly collected data and feedback data to improve the model's accuracy and performance.

[0850] This invention is a system that collects local information data from local governments and individual businesses and recommends the most suitable facilities and businesses based on user requests. This system operates based on the interaction of a server, terminals, and users.

[0851] Collection and storage of local data

[0852] The server receives local information data sent from local governments and individual businesses. This data includes facility names, addresses, contact information, and event information. The received data is analyzed by the server and classified into categories such as "facilities," "events," and "restaurants." The classified data is stored in a database (e.g., MySQL or PostgreSQL). This data forms the basis for generating recommendation information in response to user requests.

[0853] Data preprocessing and AI model training

[0854] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is converted into a format suitable for machine learning algorithms. For example, the Google Maps API is used to verify the accuracy of addresses to compensate for data defects. Using the formatted data, the server trains an AI model (e.g., using TensorFlow or PyTorch). After training is complete, the newly trained AI model is deployed to the production environment.

[0855] Processing user requests

[0856] Users request information about specific facilities or stores via LINE or a dedicated app. For example, they might send a request like, "Tell me your recommended ramen restaurant." The device sends this request to the server in real time. The server analyzes the received request using natural language processing technology (e.g., Google NLP API or SpaCy). It then searches the database for relevant information, and an AI model generates the most suitable recommendation information.

[0857] Providing recommendation information

[0858] The server sends the generated recommendation information to the device. For example, the user is provided with a recommendation such as, "The ramen shop in front of the station has high ratings." The device then displays this information to the user. In the case of LINE, it is displayed as a message, and in the case of a dedicated app, it is displayed as an in-app notification or on the screen.

[0859] Gathering feedback and relearning

[0860] Users provide feedback after using recommended facilities or stores. For example, they might send a simple rating such as "It was delicious." The device sends this feedback to the server. The server stores the feedback data in a database and uses it for training the next AI model. The AI ​​model, retrained with the new feedback data, generates even more accurate recommendations.

[0861] Specific example

[0862] A user sends a request via the LINE app saying, "Tell me your recommended ramen restaurant." The device sends this request to the server. The server searches its database for information on relevant ramen restaurants and uses an AI model to make the best recommendation. For example, the user is provided with a recommendation such as, "The ramen restaurant '○○' near the station has high ratings."

[0863] By presenting specific situations in this way, the generative AI model becomes more likely to generate specific and appropriate flows.

[0864] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0865] Step 1: Collection and reception of local information data

[0866] The server receives local information data sent from local governments and individual businesses via APIs and database connections. Specifically, the server uses API crawling and scraping techniques to retrieve data such as facility names, addresses, contact information, and event information. The input is data from local governments and individual businesses, and the output is raw local information data stored in the server's internal database.

[0867] Step 2: Analysis and classification of regional information data

[0868] The server analyzes the received data using natural language processing techniques (e.g., Google NLP API or SpaCy). Specifically, it performs text analysis and keyword matching to classify the data into categories such as "facilities," "events," and "restaurants." The input is the raw data received in step 1, and the output is the data categorized.

[0869] Step 3: Save data

[0870] The server stores the classified data in a relational database (e.g., MySQL or PostgreSQL). Specifically, the server uses SQL queries to store the data in the appropriate tables within the database. The input is the data classified in step 2, and the output is the structured data stored in the database.

[0871] Step 4: Data Cleansing

[0872] The server periodically checks the data in the database, corrects incomplete data, and removes noisy data. Specifically, it uses the Google Maps API to fill in missing address information and standardizes data with inconsistent formats. The input is existing data in the database, and the output is cleansed and formatted data.

[0873] Step 5: Training the AI ​​model

[0874] The server uses the cleansed data to train an AI model using machine learning algorithms (e.g., TensorFlow or PyTorch). Specifically, it processes large amounts of data on a GPU (e.g., NVIDIA Tesla V100) and learns patterns to generate the expected output. The input is the cleansed data, and the output is the trained AI model.

[0875] Step 6: Deploying the Artificial Intelligence Model

[0876] The server deploys the trained AI model to the production environment. Specifically, it uploads the model to the server and makes it accessible via API. The input is the trained AI model, and the output is the publicly available AI model.

[0877] Step 7: Receiving and parsing user requests

[0878] Users request information about specific facilities or stores via LINE or a dedicated application. The device sends this request to the server in real time. The server analyzes the received request using natural language processing technology and converts it into structured data. The input is the user request, and the output is the analyzed request data.

[0879] Step 8: Generating and searching for recommendation information

[0880] The server searches the database for relevant information based on the parsed request and generates optimal recommendations using an AI model. Specifically, the server uses collaborative filtering and content-based filtering to recommend the most suitable facilities and stores to the user. The input is the parsed request data and relevant data in the database, and the output is the recommendations.

[0881] Step 9: Submitting and displaying recommendation information

[0882] The server sends the generated recommendation information to the device. The device then displays this information to the user. Specifically, in the case of LINE, it is provided to the user as a message, and in the case of a dedicated app, it is provided as an in-app notification or screen display. The input is the recommendation information, and the output is the recommendation content provided to the user.

[0883] Step 10: Gathering Feedback and Retraining

[0884] Users provide feedback after using recommended facilities or stores. Specifically, users send ratings and comments via a dedicated app or LINE. The device sends this feedback to a server. The server stores the feedback data in a database and retrains the AI ​​model by incorporating the new data. The input is the user's feedback, and the output is the retrained AI model.

[0885] (Application Example 1)

[0886] Next, we will explain Application Example 1. In the following explanation, 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."

[0887] Traditional facility and store recommendation systems suffer from insufficient collection and analysis of local information, making it difficult to respond quickly and appropriately to user requests. Furthermore, the inability to provide personalized recommendations based on user preferences and usage history makes improving user satisfaction a challenge. Additionally, the inability to offer value-added services such as real-time event notifications makes it difficult to maintain user interest.

[0888] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0889] In this invention, the server includes means for receiving local information data from local governments and individual stores; means for classifying the received local information data and storing it in a database; means for training an artificial intelligence model to generate facility and store recommendations based on user requests using the stored local information data; means for receiving requests from users; means for searching the database for relevant information based on user requests and generating recommendations using the artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; means for receiving feedback from users, storing the received feedback in a database and using it for retraining; means for processing user requests via an application installed on a smartphone; and means for providing information on recommended stores and real-time event notifications via push notifications. This enables personalized recommendations of facilities and stores tailored to the user's interests and preferences, and improves the user experience through real-time information provision and feedback reflection.

[0890] A "local government" is a public institution that carries out administrative duties in a specific region and is a source of regional information data.

[0891] A "personal store" is a small-scale commercial facility or service provider owned or operated by an individual, and is a source of local information data.

[0892] "Local information data" refers to information about facilities, shops, events, and services located within a specific area, and is provided by local governments and individual businesses.

[0893] A "database" is a digital system for organizing and storing regional information data in a searchable and accessible format.

[0894] An "artificial intelligence model" is an algorithm or computational model that learns from a large amount of data and performs a specific task; in this context, it refers to a model used in recommendation systems.

[0895] A "user terminal" refers to a device used by a user, such as a computer, smartphone, or tablet, which is used to receive recommendation information and send requests.

[0896] "Feedback" refers to opinions and impressions, such as ratings and comments, provided by users, and is information used for system improvement and retraining.

[0897] "Retraining" is the process of updating an existing artificial intelligence model using newly collected data and feedback to improve its performance and accuracy.

[0898] A "smartphone" is a portable computer with mobile communication and internet connectivity capabilities, and is a device that allows for the installation and use of applications.

[0899] "Push notifications" are notification messages automatically sent from a server to a user's device, providing a means of delivering important information and updates in real time.

[0900] A "user interface" is an interface through which a user interacts with a system, providing information using messaging applications and push notification functions.

[0901] To implement this invention, interaction between a server, a terminal, and a user is necessary. A detailed embodiment is shown below.

[0902] Collection and storage of local data

[0903] First, the server receives local information data from local governments and individual businesses. This data is categorized into categories such as "facilities," "events," and "restaurants." The categorized data is stored in a database such as Firebase. This data forms the basis for generating recommendation information in response to user requests.

[0904] Data preprocessing and AI model training

[0905] Next, the server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is then formatted and converted into a format suitable for machine learning algorithms. Using this formatted data, an AI model is trained using Python libraries such as Scikit-learn. The newly trained AI model is deployed to the production environment as needed. This enables the provision of highly accurate recommendation information to users.

[0906] Processing user requests

[0907] Users can request information about facilities and shops through applications or messaging apps installed on their smartphones. In this case, if a user sends a request such as "Tell me a good ramen shop," the device sends this request to the server. The server analyzes the received request and searches its database for relevant information. Based on the retrieved information, an AI model generates the most suitable recommendations for the user. The generated recommendations are then sent as push notifications to the user's smartphone, for example, "The ramen shop in front of the station has high ratings."

[0908] Gathering feedback and relearning

[0909] After a user visits a recommended facility or store, the terminal sends feedback to the server. For example, a comment such as "It was delicious" might be sent. The server stores this feedback data in a database and uses it for retraining. This new feedback data is also integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[0910] Specific example

[0911] As a concrete example, when a user sends a request using a smartphone application saying, "Tell me some recommended cafes," the server searches for relevant cafes in real time, and the AI ​​model can recommend cafes with high ratings. The user is notified of the recommendation, such as, "The cafe near the station has high ratings." After the user visits the cafe, they can send feedback such as, "It was very comfortable," and this information will be used to train the AI ​​model for the next time.

[0912] This system will allow users to receive recommendations for facilities and shops that best suit their interests and preferences, and is expected to further stimulate local consumer activity.

[0913] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0914] Step 1:

[0915] The server receives local information data from local governments and individual businesses and stores it in a database. Specifically, it retrieves data via CSV files or APIs and stores it in the database as structured data. The input is local information data, and the output is organized and categorized data.

[0916] Step 2:

[0917] The server periodically checks the regional information data in the database, corrects incomplete data, and removes noisy data. For example, it imputes missing values ​​and removes inappropriate values. This process results in cleansed data. The input is the regional information data in the database, and the output is the cleansed data.

[0918] Step 3:

[0919] The server uses the cleansed data to transform it into a format suitable for machine learning algorithms and trains an AI model. It uses Python's Scikit-learn to vectorize the data and train the model. The input is the cleansed data, and the output is the trained AI model.

[0920] Step 4:

[0921] Users request information about facilities and stores through a smartphone application. For example, they might type "Tell me your recommended ramen restaurant" as text. The input is the user's request, and the output is the request data.

[0922] Step 5:

[0923] The terminal sends the user's request to the server. The server parses the request and searches the database for relevant information. For example, it might perform a search based on the keyword "ramen shop." The input is the request data, and the output is the relevant information.

[0924] Step 6:

[0925] The server generates optimal recommendations using an AI model based on the searched information. It uses a trained model to select highly-rated stores. The input is relevant information, and the output is recommendations.

[0926] Step 7:

[0927] The server sends the generated recommendation information to the user's terminal. For example, it might send a push notification saying, "The ramen shop in front of the station has high ratings." The input is the recommendation information, and the output is the notification sent to the user's terminal.

[0928] Step 8:

[0929] Users visit recommended facilities or shops and then provide feedback. They submit comments such as "It was delicious." The input is the user's feedback, and the output is feedback data.

[0930] Step 9:

[0931] The device sends user feedback to the server. The server stores the feedback data in a database and uses it for retraining. The collected feedback data is also integrated with existing data and used to retrain the AI ​​model. The input is the feedback data, and the output is the updated AI model.

[0932] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0933] This invention is a system for collecting local information data from local governments and individual businesses, recognizing user emotions using an emotion engine based on user requests and feedback, and utilizing that information to recommend the most suitable facilities and businesses. This system operates based on the interaction between a server, a terminal, and a user.

[0934] Collection and storage of local data

[0935] The server receives local information data sent from local governments and individual businesses. The received data is appropriately analyzed, categorized into categories such as "facilities," "events," and "restaurants," and stored in a database. This data forms the basis for generating recommendation information in response to user requests.

[0936] Data preprocessing and AI model training

[0937] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is formatted and converted into a format suitable for machine learning algorithms. This formatted data is used to train the AI ​​model. The newly trained AI model is deployed to the production environment in a timely manner. This enables the provision of highly accurate recommendation information to users.

[0938] Processing user requests and sentiment analysis

[0939] Users request information about facilities and stores through LINE or other applications. The device sends this request to the server. The server analyzes the received request and understands its content. Furthermore, it can use an emotion engine to extract emotional information from the user's request. For example, if a user requests "I'm hungry and very irritated," the emotion engine recognizes "irritated."

[0940] Generation and provision of recommendation information

[0941] The server searches the database for information related to the request. Furthermore, using the retrieved sentiment information, an AI model generates recommendations best suited to the user. For example, a user who is "frustrated" might be recommended a restaurant that can provide service quickly. This recommendation information is then delivered to the user via their device.

[0942] As a concrete example, a user requests via the LINE app, "Tell me about a good ramen restaurant quickly." The device sends this request to the server. The server analyzes the request and uses an emotion engine to obtain "urgent" emotion information. The server searches its database for ramen restaurants that can respond quickly and uses an AI model to recommend, "The ramen restaurant 'Mensho' near the station is highly rated for its quick service." The device then provides this information to the user.

[0943] Gathering feedback and relearning

[0944] After a user visits a recommended facility or store, they provide feedback. The device sends this feedback to a server. The server stores the feedback data in a database and uses it for retraining. The sentiment information included in the provided feedback is also analyzed and stored in the database. This new feedback data is integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[0945] As a concrete example, a user might submit feedback about a ramen restaurant they visited, stating, "It was very delicious, but the service was a little slow." This feedback is sent to the server and stored in the database. The server uses this feedback data and the sentiment information analyzed by the sentiment engine to retrain the AI ​​model and incorporate it into future recommendations.

[0946] Thus, the present invention is a system that efficiently collects local information from local governments and individual shops, analyzes users' emotions using an emotion engine, and provides users with highly accurate recommendation information, thereby stimulating local consumer activity and improving their dwell time.

[0947] The following describes the processing flow.

[0948] Collection and storage of local data

[0949] Step 1:

[0950] The server receives HTTP POST requests from local governments and individual businesses.

[0951] Step 2:

[0952] The server parses the received request body and extracts regional information data.

[0953] Step 3:

[0954] The server categorizes the extracted data into categories such as "facilities," "events," and "restaurants."

[0955] Step 4:

[0956] The server stores the classified data in a database.

[0957] Data preprocessing and AI model training

[0958] Step 1:

[0959] The server periodically checks the raw data in the database, corrects incomplete data, and removes noisy data.

[0960] Step 2:

[0961] The server formats the cleansed data into a format suitable for machine learning algorithms.

[0962] Step 3:

[0963] The server uses the formatted data to train the AI ​​model.

[0964] Step 4:

[0965] The server evaluates the model's performance and adjusts parameters as needed.

[0966] Step 5:

[0967] The server uploads the newly trained model to the production deployment environment.

[0968] Processing user requests and sentiment analysis

[0969] Step 1:

[0970] Users request information about facilities and stores through LINE or other apps.

[0971] Step 2:

[0972] The terminal sends the user's request to the server as an HTTP request.

[0973] Step 3:

[0974] The server analyzes the received request to understand its content and the user's intent.

[0975] Step 4:

[0976] The server uses an emotion engine to extract emotional information from user requests.

[0977] Step 5:

[0978] The server searches the database for information related to the request.

[0979] Step 6:

[0980] The server uses an AI model to generate optimal recommendations based on search results and sentiment information.

[0981] Step 7:

[0982] The server formats the generated recommendation information and sends it to the terminal.

[0983] Step 8:

[0984] The device displays recommendation information on the user interface.

[0985] Specific example

[0986] Step 1:

[0987] The user sends a request on the LINE app saying "Please tell me a delicious ramen restaurant quickly."

[0988] Step 2:

[0989] The terminal sends this request to the server as an HTTP request.

[0990] Step 3:

[0991] The server analyzes the request and understands that the user is "in a hurry."

[0992] Step 4:

[0993] The server uses an emotion engine to extract the "in a hurry" emotion from the text of the request.

[0994] Step 5:

[0995] The server quickly searches the database for information on ramen restaurants that can provide services.

[0996] Step 6:

[0997] The server generates optimal recommendation information using an AI model based on the search results and emotion information.

[0998] Step 7:

[0999] The server formalizes the recommendation information "The ramen restaurant 'Noodle Master' in front of the station is evaluated for quick service" and sends it to the terminal.

[1000] Step 8:

[1001] The terminal displays this information to the user on the LINE app.

[1002] Collection and Relearning of Feedback

[1003] Step 1:

[1004] Users provide feedback after using the recommended facilities or stores.

[1005] Step 2:

[1006] The device sends feedback to the server.

[1007] Step 3:

[1008] The server analyzes the feedback and extracts emotional information.

[1009] Step 4:

[1010] The server stores feedback data and emotional information in a database.

[1011] Step 5:

[1012] The server retrains the AI ​​model using the stored feedback data.

[1013] In this way, the system achieves highly accurate information provision and recommendations tailored to user sentiment through an efficient process of collecting, storing, analyzing, learning, recommending, and providing feedback on local information.

[1014] (Example 2)

[1015] Next, we will describe Example 2. 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."

[1016] Current regional information recommendation systems have a problem in that they struggle to make recommendations that take into account user emotions and urgency. Furthermore, the presence of data incompleteness and noise often leads to decreased recommendation accuracy. Additionally, there is a lack of effective means to utilize user feedback for retraining, which hinders the continuous improvement of system performance.

[1017] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1018] In this invention, the server includes means for receiving local information data from local governments and individual stores; means for classifying the received local information data and storing it in a database; means for checking the integrity of the stored local information data and correcting and deleting incomplete or noisy data; means for converting the formatted data into a format suitable for machine learning algorithms and training an artificial intelligence model; means for receiving requests from users; means for analyzing the content of requests and obtaining sentiment information using a sentiment engine; means for searching for relevant information from the database and generating recommendation information using an artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; means for receiving feedback from users, storing the received feedback in a database and using it for retraining; and means for optimizing the recommendation information based on the generated sentiment information. This makes it possible to provide highly accurate recommendation information based on the user's sentiment and urgency.

[1019] A "local government" refers to a public institution responsible for managing and operating a specific region.

[1020] "Individual store" refers to a commercial or service facility operated by an individual or a small business.

[1021] "Local information data" refers to data that includes information about facilities, events, restaurants, etc., related to a specific region.

[1022] A "database" refers to a system for systematically storing and managing collected information.

[1023] "Incomplete data" refers to data that is missing necessary information or contains errors.

[1024] "Noise data" refers to unwanted data that hinders analysis and learning.

[1025] A "machine learning algorithm" refers to a method or model for automatically learning patterns and rules from data.

[1026] An "artificial intelligence model" refers to a program that has been trained using machine learning algorithms and possesses the ability to perform specific tasks.

[1027] An "emotion engine" refers to a technology that uses natural language processing and text mining to recognize and analyze emotions from text.

[1028] A "user terminal" refers to an electronic device (e.g., smartphone, tablet, personal computer) used by a user to input and receive information.

[1029] A "request" refers to a request made by a user to a system for the provision of information.

[1030] "Feedback" refers to the evaluations and opinions that users give regarding the information and services provided.

[1031] "Retraining" refers to the process of retraining an artificial intelligence model using new data and feedback.

[1032] "Optimization" refers to adjusting the parameters and processes of a system to their optimal state in order to achieve a specific objective.

[1033] This invention relates to a system that collects local information data from local governments and individual businesses and generates recommendation information based on user requests and feedback. This system operates based on the interaction between a server, a terminal, and a user.

[1034] Collection and storage of local data

[1035] The server receives local information data sent from local governments and individual businesses. This information is retrieved through interfaces such as APIs. Examples include event information from local governments and business hours from individual businesses. This data is categorized into categories such as "facilities," "events," and "restaurants," and stored in a database. Relational databases such as MySQL and PostgreSQL can be used as the database. This data later forms the basis for generating recommendation information in response to user requests.

[1036] Data preprocessing and AI model training

[1037] The server periodically checks the received regional data, correcting and removing incomplete and noisy data. Specifically, it fills in missing information and removes outliers. Data cleansing scripts using Python or Pandas are used. The formatted data is converted into a format suitable for machine learning algorithms. For example, categorical data is encoded into numerical data. This is then used to train the AI ​​model. TensorFlow and PyTorch are used as machine learning frameworks. The newly trained AI model is deployed to the production environment as needed.

[1038] Processing user requests and sentiment analysis

[1039] Users request information about facilities and stores through LINE or a dedicated application. For example, they might send a message like, "Tell me about a good ramen restaurant quickly." The device sends this request to the server. The server analyzes the received request and uses natural language processing technology (e.g., spaCy or NLTK) to understand its content. Furthermore, it uses an emotion engine (e.g., Google Cloud Natural Language API) to extract emotional information from the user's request. For example, it recognizes "irritated" from a request like, "I'm hungry and very irritated."

[1040] Generation and provision of recommendation information

[1041] The server searches the database for information related to the request, and the AI ​​model uses the retrieved sentiment information to generate the most suitable recommendations for the user. For example, for a user who is "frustrated," it recommends restaurants that can provide quick service. This recommendation information is provided to the user through their device. As a specific example, if a user requests "Tell me a good ramen shop quickly" using the LINE app, the device sends the request to the server, and the server analyzes the request and sentiment information. It searches the database for ramen shops that can respond quickly and uses the AI ​​model to recommend "Ramen shop 'XX' near the station is highly rated for its quick service." This information is provided to the user through their device.

[1042] Gathering feedback and relearning

[1043] Users provide feedback after using recommended facilities or stores. For example, they might send feedback via the LINE app, such as, "It was delicious, but the service was a little slow." The device sends this feedback to a server. The server stores the received feedback data in a database and uses it for retraining. The collected feedback data undergoes cleansing and formatting steps, is integrated with existing data, and then used to retrain the AI ​​model. This ensures that the system can always provide highly accurate recommendation information.

[1044] Examples of prompt statements

[1045] The following are specific examples of input prompts for a generative AI model.

[1046] 1. "Can you recommend a good cafe nearby?"

[1047] 2. "I'm looking for a restaurant that's safe and comfortable for families with children."

[1048] 3. "What are some recommended events to go to with friends?"

[1049] 4. "Please recommend a good place for a date."

[1050] By utilizing these prompts, the generative AI model can provide optimal recommendation information based on user requests.

[1051] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1052] Step 1: Collecting local data

[1053] The server receives local information data from local governments and individual businesses. The specific operation is as follows: First, input processing is performed to acquire data through an interface such as an API. Next, the received data is temporarily stored in a buffer. Then, it is classified into categories such as "facilities," "events," and "restaurants," and finally stored in the database. In this step, the input is the received local information data, and the output is the classified and stored data.

[1054] Step 2: Data preprocessing

[1055] The server periodically checks the received regional information data, correcting and removing incomplete and noisy data. Specifically, it performs data integrity checks, supplements missing information, and removes outliers. Next, it formats this data into a format suitable for machine learning algorithms. For example, it encodes categorical data into numerical data. In this step, the input is classified data, and the output is formatted data.

[1056] Step 3: Training the AI ​​model

[1057] The server trains an AI model using pre-processed data. The specific operation is as follows: First, the new data is split into a training set and a test set. Next, the AI ​​model is trained using a machine learning framework such as TensorFlow or PyTorch. The trained model is finally deployed to the production environment. The input in this step is a formatted dataset, and the output is the trained AI model.

[1058] Step 4: Processing User Requests

[1059] Users send requests via LINE or a dedicated application. A concrete example of such a request might be, "Tell me a good ramen restaurant quickly." The device sends this request to the server. The server uses natural language processing techniques to analyze the received request and understand its content. In this step, the input is the user request, and the output is the analyzed request content.

[1060] Step 5: Emotion Analysis

[1061] The server analyzes the received request using an emotion engine to obtain the user's emotion information. Specifically, it uses a natural language processing library to extract keywords and emotional expressions from the request and obtains the emotion information through the emotion engine. For example, from the request "Tell me a good ramen restaurant quickly," the emotion information "I'm in a hurry" is extracted. In this step, the input is the analyzed request content, and the output is the obtained emotion information.

[1062] Step 6: Generating recommendation information

[1063] The server searches the database for relevant information and generates optimal recommendations based on the retrieved sentiment information and user requests. Specifically, it uses an AI model to evaluate the request content and sentiment data to generate optimal recommendations. For example, it might recommend stores that can provide quick service to a user who is "in a hurry." The inputs in this step are sentiment information and user requests, and the output is the generated recommendations.

[1064] Step 7: Providing Recommendation Information

[1065] The server provides the generated recommendation information to the user via the terminal. Specifically, the recommendation information is sent to the terminal via an API, and the terminal displays that information to the user. For example, a recommendation such as "The ramen shop 'XX' in front of the station is highly rated for its fast service" is notified to the LINE app. In this step, the input is the generated recommendation information, and the output is the notification to the user.

[1066] Step 8: Gathering Feedback

[1067] Users provide feedback after using a facility or store. For example, they might say, "The food was delicious, but the service was a little slow." The device sends this feedback to the server, which then stores it in a database. In this step, the input is the user feedback, and the output is the stored feedback data.

[1068] Step 9: Retraining the AI ​​model

[1069] The server retrains the AI ​​model using the collected feedback data. Specifically, it cleanses and formats the new feedback data and integrates it with the existing dataset. Then, it retrains the AI ​​model and deploys it to the production environment. In this step, the input is the saved feedback data, and the output is the retrained AI model.

[1070] (Application Example 2)

[1071] Next, we will explain application example 2. In the following explanation, 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."

[1072] Traditional local information recommendation systems have a problem of not providing sufficient user satisfaction because they simply provide information without considering the user's emotional state. Furthermore, online virtual shopping lacks personalized recommendations based on user emotions and feedback. This results in a uniform shopping experience for all users, making it difficult to recommend optimal products and services that meet individual needs.

[1073] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for receiving local information data from local governments and individual stores; means for classifying the received local information data and storing it in a database; means for training an artificial intelligence model for generating facility and store recommendations based on user requests using the stored local information data; means for searching for relevant information from the database based on user requests and sentiment information and generating recommendations using the artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; means for receiving user feedback, storing the received feedback in a database and using it for retraining; and means for recommending products and services in the virtual store based on the stored local information data and user sentiment information. This makes it possible to provide detailed information recommendations based on user sentiment and feedback, thereby improving user satisfaction and the shopping experience.

[1074] A "local government" is an administrative body that governs a specific region and is responsible for providing public services and promoting the development of the local community.

[1075] A "private store" is a commercial facility owned and operated by an individual, primarily providing goods or services.

[1076] "Local information data" refers to data provided by local governments or individual businesses that includes various types of information about a particular region.

[1077] A "database" is a system for efficiently storing, searching, and managing data.

[1078] "Emotional information" refers to data that indicates a user's emotional state, and is primarily analyzed by an emotion engine.

[1079] A "user terminal" is a device used by a user to interact with the system, and includes smartphones, tablets, and other similar devices.

[1080] An "artificial intelligence model" is an algorithm or system that automatically performs a specific task by analyzing and learning from data.

[1081] "Feedback" refers to data, including user opinions and impressions after using a system, which is used to improve the system.

[1082] A "virtual store" is a commercial facility in an online environment that provides goods and services via the internet.

[1083] "Recommendation information" refers to information about the most suitable facilities and stores, generated based on user requests and sentiments.

[1084] "Retraining" is the process of retraining an existing artificial intelligence model using new data and feedback.

[1085] This invention is a system that collects local information data and generates and provides recommendation information based on user requests and sentiment information. Specific embodiments of this invention are described below.

[1086] 1. Data collection and storage

[1087] The server receives local information data from local governments and individual businesses. This data includes a wide range of information such as facilities, events, and restaurants. The received data is appropriately analyzed, categorized, and then stored in a database.

[1088] 2. Data preprocessing and AI model training

[1089] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is formatted to a format suitable for machine learning algorithms, and this is used to train AI models. The newly trained AI models are deployed to the production environment, enabling them to provide users with highly accurate recommendations.

[1090] 3. Processing User Requests and Sentiment Analysis

[1091] Users request information through the virtual store's shopping app or messaging application. The device sends this request to the server. The server analyzes the received request and uses an emotion engine to obtain the user's emotional information. For example, if a user requests, "I'm feeling stressed today, so I'm looking for a relaxing cafe. Do you have any recommendations?", the emotion engine recognizes "stress."

[1092] 4. Generation and provision of recommendation information

[1093] The server searches the database for information related to the request, and using the retrieved sentiment information, the AI ​​model generates recommendations best suited to the user. For example, a user feeling "stressed" would be recommended relaxing cafes or shops. This recommendation information is then sent to the user's device.

[1094] 5. Gathering feedback and relearning

[1095] After a user uses a recommended facility or store, they provide feedback. The device sends this feedback to a server. The server stores the feedback data in a database and uses it for retraining. The sentiment information included in the provided feedback is also analyzed and stored in the database. This new feedback data is integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[1096] Hardware and software to be used

[1097] Hardware: Servers, smartphones, tablets

[1098] Software: Web frameworks (e.g., Flask, Django), machine learning libraries (e.g., TensorFlow), sentiment analysis engines (e.g., Emotion-Recognition-API)

[1099] Specific example

[1100] When a user requests, "I'm feeling stressed today, so I'm looking for a relaxing cafe. Do you have any recommendations?", the emotion engine recognizes the user's "stress" and recommends a relaxing establishment. Similarly, if a user requests, "Tell me about a good ramen shop quickly," the system can recommend a ramen shop that can provide quick service based on the emotional information that the user is "in a hurry."

[1101] Example of a prompt:

[1102] "I'm feeling stressed today, so I'm looking for a relaxing cafe. Do you have any recommendations?"

[1103] "Please tell me a good ramen shop quickly. Thank you."

[1104] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1105] Step 1:

[1106] The server receives local information data from local governments and individual businesses. This data is often provided in a structured format such as JSON. The received data is saved to a database in real time. The input is local information data sent from local governments and individual businesses, and the output is the saved data.

[1107] Step 2:

[1108] The server classifies the data in the database and stores it by category. For example, it might classify data into categories such as "restaurants," "events," and "facilities." This classification is performed using a data analysis algorithm. The input is the stored regional information data, and the output is the classified data.

[1109] Step 3:

[1110] The server uses the cleansed data to train an artificial intelligence model. Data shaping processes convert the data into a format suitable for machine learning algorithms, and this is then input into the AI ​​model. This dataset is used to train the model and build a highly accurate recommendation engine. The input is the shaped data, and the output is the trained AI model.

[1111] Step 4:

[1112] The user sends a request via a smartphone or tablet. The request content is parsed using natural language processing. The device sends this request to the server, and the prompt text includes the user's sentiment. The input is the user request, and the output is the parsing result.

[1113] Step 5:

[1114] The server analyzes the received request and uses an emotion engine to obtain the user's emotional information. For example, the emotion engine analyzes emotional information such as "I want to relax," "I'm in a hurry," or "I'm stressed." The input is the user's request, and the output is the user's emotional information.

[1115] Step 6:

[1116] The server searches the database for relevant information and uses an AI model to generate optimal recommendations based on the retrieved sentiment information. For example, based on sentiment information such as "I want to relax," it might recommend quiet cafes or parks. The input is the user's sentiment information and request, and the output is the recommendations.

[1117] Step 7:

[1118] The server sends the generated recommendation information to the user's terminal. The user can receive this information on their smartphone or tablet. The input is recommendation information, and the output is visualized recommendation information.

[1119] Step 8:

[1120] Users utilize a facility or store and then provide feedback. The terminal sends this feedback to a server. The input is the user's feedback, and the output is the analyzed feedback information.

[1121] Step 9:

[1122] The server stores the feedback it receives in a database and uses it for retraining. This allows the AI ​​model to continuously learn and improve its recommendation accuracy. The input is the analyzed feedback information, and the output is the updated AI model.

[1123] The above outlines the specific processing steps of the system that implements the application example.

[1124] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1125] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1126] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1127] [Fourth Embodiment]

[1128] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1129] As shown in Figure 7, the 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.

[1130] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1131] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1132] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1133] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1134] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1135] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1136] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1137] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[1138] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1139] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1140] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1141] This invention is a system for collecting local information data from local governments and individual businesses and recommending the most suitable facilities and stores based on user requests. This system operates based on the interaction between a server, a terminal, and a user.

[1142] Collection and storage of local data

[1143] The server receives local information data sent from local governments and individual businesses. The received data is appropriately analyzed and classified into categories such as "facilities," "events," and "restaurants." This classified data is stored in a database. This data forms the basis for generating recommendation information in response to user requests.

[1144] Data preprocessing and AI model training

[1145] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is formatted and converted into a format suitable for machine learning algorithms. This formatted data is used to train the AI ​​model. The newly trained AI model is deployed to the production environment in a timely manner. This enables the provision of highly accurate recommendation information to users.

[1146] Processing user requests

[1147] Users request information about facilities and stores through LINE or other applications. The device sends this request to a server. The server analyzes the received request and searches its database for relevant information. Based on the retrieved information, an AI model generates the most suitable recommendations for the user. These recommendations are then delivered to the user through their device.

[1148] As a concrete example, a user sends a request via the LINE app saying, "Tell me your recommended ramen restaurant." The device sends this request to the server. The server searches its database for information on relevant ramen restaurants and uses an AI model to make the best recommendation. For example, the user might be provided with a recommendation such as, "The ramen restaurant 'Mensho' near the station has high ratings."

[1149] Gathering feedback and relearning

[1150] After a user uses a recommended facility or store, they provide feedback. The device sends this feedback to the server. The server stores the feedback data in a database and uses it for retraining. This new feedback data is also integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[1151] As a concrete example, a user sends feedback saying "It was very delicious" about a ramen restaurant they visited. This feedback is sent to the server and stored in the database. The server uses this feedback data to retrain the AI ​​model and incorporate it into future recommendations.

[1152] Thus, the present invention is a system that efficiently collects local information from local governments and individual shops and provides users with highly accurate recommendation information, thereby stimulating local consumer activity and improving the likelihood of people staying in the area.

[1153] The following describes the processing flow.

[1154] Collection and storage of local data

[1155] Step 1:

[1156] The server receives HTTP POST requests from local governments and individual businesses.

[1157] Step 2:

[1158] The server parses the received request body and extracts regional information data.

[1159] Step 3:

[1160] The server categorizes the extracted data into categories such as "facilities," "events," and "restaurants."

[1161] Step 4:

[1162] The server stores the classified data in a database.

[1163] Data preprocessing and AI model training

[1164] Step 1:

[1165] The server periodically checks the raw data in the database, corrects incomplete data, and removes noisy data.

[1166] Step 2:

[1167] The server formats the cleansed data into a format suitable for machine learning algorithms.

[1168] Step 3:

[1169] The server uses the formatted data to train the AI ​​model.

[1170] Step 4:

[1171] The server evaluates the model's performance and adjusts parameters as needed.

[1172] Step 5:

[1173] The server uploads the newly trained model to the production deployment environment.

[1174] Processing user requests

[1175] Step 1:

[1176] Users request information about facilities and stores through LINE or other apps.

[1177] Step 2:

[1178] The terminal sends the user's request to the server as an HTTP request.

[1179] Step 3:

[1180] The server parses the received request and understands its contents.

[1181] Step 4:

[1182] The server searches the database for information related to the request.

[1183] Step 5:

[1184] The server uses an AI model to generate optimal recommendations based on the search results.

[1185] Step 6:

[1186] The server formats the generated recommendation information and sends it to the terminal.

[1187] Step 7:

[1188] The device displays recommendation information on the user interface.

[1189] Gathering feedback and relearning

[1190] Step 1:

[1191] Users provide feedback after using the recommended facilities or stores.

[1192] Step 2:

[1193] The device sends feedback to the server.

[1194] Step 3:

[1195] The server saves the feedback to the database.

[1196] Step 4:

[1197] The server integrates the new feedback data with the existing data.

[1198] Step 5:

[1199] The server retrains the AI ​​model using the integrated data.

[1200] In this way, the system achieves highly accurate information delivery through efficient processes of collecting, storing, analyzing, learning from, recommending, and providing feedback on local information.

[1201] (Example 1)

[1202] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1203] Conventional local information recommendation systems often suffered from incomplete data collection or reduced recommendation accuracy due to the inclusion of noisy data. Furthermore, responses to user requests were sometimes not in real time, compromising usability. In addition, insufficient retraining using feedback made continuous improvement of recommendation accuracy difficult. This invention aims to effectively solve these problems and improve the recommendation accuracy for local facilities and stores.

[1204] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1205] In this invention, the server includes means for receiving local information data from local governments and individual stores; means for analyzing the received local information data, classifying it into categories such as "facilities," "events," and "restaurants," and storing it in a database; means for periodically cleansing the stored local information data to correct defects and remove noise data; means for training an artificial intelligence model using a machine learning algorithm with the cleansed data; means for deploying the trained artificial intelligence model; means for analyzing user requests using natural language processing technology; means for searching for relevant information from the database based on the analyzed requests and generating recommendation information using the artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; and means for receiving user feedback, storing the received feedback in a database, and using it for retraining. This improves the accuracy of local information data collection and cleansing, and enables real-time responses. Furthermore, the recommendation accuracy continuously improves by incorporating feedback.

[1206] "Local information data" refers to various types of information provided by local governments and individual businesses, such as facility names, addresses, contact information, and event information.

[1207] "Analysis" refers to the process of appropriately assigning and classifying received data into each category.

[1208] A "category" refers to a specific group used to classify information such as "facilities," "events," and "restaurants."

[1209] A "database" refers to a data management system for efficiently storing, searching, and updating information.

[1210] "Cleansing" refers to the process of correcting data defects and removing noisy data.

[1211] A "machine learning algorithm" refers to a mathematical model that uses large amounts of data to detect patterns and makes predictions and classifications based on new data.

[1212] An "artificial intelligence model" refers to a system that uses machine learning algorithms to analyze data, make predictions, and generate recommendation information.

[1213] "Deployment" refers to introducing a trained artificial intelligence model into a production environment and putting it into operation.

[1214] "Natural language processing technology" refers to computer science techniques used to analyze human language and understand its meaning.

[1215] "Recommendation information" refers to information about the most suitable facilities and stores, generated based on user requests.

[1216] "Feedback" refers to information such as ratings and comments that users provide after using a recommended facility or store.

[1217] "Retraining" refers to the process of retraining an artificial intelligence model using newly collected data and feedback data to improve the model's accuracy and performance.

[1218] This invention is a system that collects local information data from local governments and individual businesses and recommends the most suitable facilities and businesses based on user requests. This system operates based on the interaction of a server, terminals, and users.

[1219] Collection and storage of local data

[1220] The server receives local information data sent from local governments and individual businesses. This data includes facility names, addresses, contact information, and event information. The received data is analyzed by the server and classified into categories such as "facilities," "events," and "restaurants." The classified data is stored in a database (e.g., MySQL or PostgreSQL). This data forms the basis for generating recommendation information in response to user requests.

[1221] Data preprocessing and AI model training

[1222] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is converted into a format suitable for machine learning algorithms. For example, the Google Maps API is used to verify the accuracy of addresses to compensate for data defects. Using the formatted data, the server trains an AI model (e.g., using TensorFlow or PyTorch). After training is complete, the newly trained AI model is deployed to the production environment.

[1223] Processing user requests

[1224] Users request information about specific facilities or stores via LINE or a dedicated app. For example, they might send a request like, "Tell me your recommended ramen restaurant." The device sends this request to the server in real time. The server analyzes the received request using natural language processing technology (e.g., Google NLP API or SpaCy). It then searches the database for relevant information, and an AI model generates the most suitable recommendation information.

[1225] Providing recommendation information

[1226] The server sends the generated recommendation information to the device. For example, the user is provided with a recommendation such as, "The ramen shop in front of the station has high ratings." The device then displays this information to the user. In the case of LINE, it is displayed as a message, and in the case of a dedicated app, it is displayed as an in-app notification or on the screen.

[1227] Gathering feedback and relearning

[1228] Users provide feedback after using recommended facilities or stores. For example, they might send a simple rating such as "It was delicious." The device sends this feedback to the server. The server stores the feedback data in a database and uses it for training the next AI model. The AI ​​model, retrained with the new feedback data, generates even more accurate recommendations.

[1229] Specific example

[1230] A user sends a request via the LINE app saying, "Tell me your recommended ramen restaurant." The device sends this request to the server. The server searches its database for information on relevant ramen restaurants and uses an AI model to make the best recommendation. For example, the user is provided with a recommendation such as, "The ramen restaurant '○○' near the station has high ratings."

[1231] By presenting specific situations in this way, the generative AI model becomes more likely to generate specific and appropriate flows.

[1232] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1233] Step 1: Collection and reception of local information data

[1234] The server receives local information data sent from local governments and individual businesses via APIs and database connections. Specifically, the server uses API crawling and scraping techniques to retrieve data such as facility names, addresses, contact information, and event information. The input is data from local governments and individual businesses, and the output is raw local information data stored in the server's internal database.

[1235] Step 2: Analysis and classification of regional information data

[1236] The server analyzes the received data using natural language processing techniques (e.g., Google NLP API or SpaCy). Specifically, it performs text analysis and keyword matching to classify the data into categories such as "facilities," "events," and "restaurants." The input is the raw data received in step 1, and the output is the data categorized.

[1237] Step 3: Save data

[1238] The server stores the classified data in a relational database (e.g., MySQL or PostgreSQL). Specifically, the server uses SQL queries to store the data in the appropriate tables within the database. The input is the data classified in step 2, and the output is the structured data stored in the database.

[1239] Step 4: Data Cleansing

[1240] The server periodically checks the data in the database, corrects incomplete data, and removes noisy data. Specifically, it uses the Google Maps API to fill in missing address information and standardizes data with inconsistent formats. The input is existing data in the database, and the output is cleansed and formatted data.

[1241] Step 5: Training the AI ​​model

[1242] The server uses the cleansed data to train an AI model using machine learning algorithms (e.g., TensorFlow or PyTorch). Specifically, it processes large amounts of data on a GPU (e.g., NVIDIA Tesla V100) and learns patterns to generate the expected output. The input is the cleansed data, and the output is the trained AI model.

[1243] Step 6: Deploying the Artificial Intelligence Model

[1244] The server deploys the trained AI model to the production environment. Specifically, it uploads the model to the server and makes it accessible via API. The input is the trained AI model, and the output is the publicly available AI model.

[1245] Step 7: Receiving and parsing user requests

[1246] Users request information about specific facilities or stores via LINE or a dedicated application. The device sends this request to the server in real time. The server analyzes the received request using natural language processing technology and converts it into structured data. The input is the user request, and the output is the analyzed request data.

[1247] Step 8: Generating and searching for recommendation information

[1248] The server searches the database for relevant information based on the parsed request and generates optimal recommendations using an AI model. Specifically, the server uses collaborative filtering and content-based filtering to recommend the most suitable facilities and stores to the user. The input is the parsed request data and relevant data in the database, and the output is the recommendations.

[1249] Step 9: Submitting and displaying recommendation information

[1250] The server sends the generated recommendation information to the device. The device then displays this information to the user. Specifically, in the case of LINE, it is provided to the user as a message, and in the case of a dedicated app, it is provided as an in-app notification or screen display. The input is the recommendation information, and the output is the recommendation content provided to the user.

[1251] Step 10: Gathering Feedback and Retraining

[1252] Users provide feedback after using recommended facilities or stores. Specifically, users send ratings and comments via a dedicated app or LINE. The device sends this feedback to a server. The server stores the feedback data in a database and retrains the AI ​​model by incorporating the new data. The input is the user's feedback, and the output is the retrained AI model.

[1253] (Application Example 1)

[1254] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1255] Traditional facility and store recommendation systems suffer from insufficient collection and analysis of local information, making it difficult to respond quickly and appropriately to user requests. Furthermore, the inability to provide personalized recommendations based on user preferences and usage history makes improving user satisfaction a challenge. Additionally, the inability to offer value-added services such as real-time event notifications makes it difficult to maintain user interest.

[1256] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1257] In this invention, the server includes means for receiving local information data from local governments and individual stores; means for classifying the received local information data and storing it in a database; means for training an artificial intelligence model to generate facility and store recommendations based on user requests using the stored local information data; means for receiving requests from users; means for searching the database for relevant information based on user requests and generating recommendations using the artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; means for receiving feedback from users, storing the received feedback in a database and using it for retraining; means for processing user requests via an application installed on a smartphone; and means for providing information on recommended stores and real-time event notifications via push notifications. This enables personalized recommendations of facilities and stores tailored to the user's interests and preferences, and improves the user experience through real-time information provision and feedback reflection.

[1258] A "local government" is a public institution that carries out administrative duties in a specific region and is a source of regional information data.

[1259] A "personal store" is a small-scale commercial facility or service provider owned or operated by an individual, and is a source of local information data.

[1260] "Local information data" refers to information about facilities, shops, events, and services located within a specific area, and is provided by local governments and individual businesses.

[1261] A "database" is a digital system for organizing and storing regional information data in a searchable and accessible format.

[1262] An "artificial intelligence model" is an algorithm or computational model that learns from a large amount of data and performs a specific task; in this context, it refers to a model used in recommendation systems.

[1263] A "user terminal" refers to a device used by a user, such as a computer, smartphone, or tablet, which is used to receive recommendation information and send requests.

[1264] "Feedback" refers to opinions and impressions, such as ratings and comments, provided by users, and is information used for system improvement and retraining.

[1265] "Retraining" is the process of updating an existing artificial intelligence model using newly collected data and feedback to improve its performance and accuracy.

[1266] A "smartphone" is a portable computer with mobile communication and internet connectivity capabilities, and is a device that allows for the installation and use of applications.

[1267] "Push notifications" are notification messages automatically sent from a server to a user's device, providing a means of delivering important information and updates in real time.

[1268] A "user interface" is an interface through which a user interacts with a system, providing information using messaging applications and push notification functions.

[1269] To implement this invention, interaction between a server, a terminal, and a user is necessary. A detailed embodiment is shown below.

[1270] Collection and storage of local data

[1271] First, the server receives local information data from local governments and individual businesses. This data is categorized into categories such as "facilities," "events," and "restaurants." The categorized data is stored in a database such as Firebase. This data forms the basis for generating recommendation information in response to user requests.

[1272] Data preprocessing and AI model training

[1273] Next, the server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is then formatted and converted into a format suitable for machine learning algorithms. Using this formatted data, an AI model is trained using Python libraries such as Scikit-learn. The newly trained AI model is deployed to the production environment as needed. This enables the provision of highly accurate recommendation information to users.

[1274] Processing user requests

[1275] Users can request information about facilities and shops through applications or messaging apps installed on their smartphones. In this case, if a user sends a request such as "Tell me a good ramen shop," the device sends this request to the server. The server analyzes the received request and searches its database for relevant information. Based on the retrieved information, an AI model generates the most suitable recommendations for the user. The generated recommendations are then sent as push notifications to the user's smartphone, for example, "The ramen shop in front of the station has high ratings."

[1276] Gathering feedback and relearning

[1277] After a user visits a recommended facility or store, the terminal sends feedback to the server. For example, a comment such as "It was delicious" might be sent. The server stores this feedback data in a database and uses it for retraining. This new feedback data is also integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[1278] Specific example

[1279] As a concrete example, when a user sends a request using a smartphone application saying, "Tell me some recommended cafes," the server searches for relevant cafes in real time, and the AI ​​model can recommend cafes with high ratings. The user is notified of the recommendation, such as, "The cafe near the station has high ratings." After the user visits the cafe, they can send feedback such as, "It was very comfortable," and this information will be used to train the AI ​​model for the next time.

[1280] This system will allow users to receive recommendations for facilities and shops that best suit their interests and preferences, and is expected to further stimulate local consumer activity.

[1281] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1282] Step 1:

[1283] The server receives local information data from local governments and individual businesses and stores it in a database. Specifically, it retrieves data via CSV files or APIs and stores it in the database as structured data. The input is local information data, and the output is organized and categorized data.

[1284] Step 2:

[1285] The server periodically checks the regional information data in the database, corrects incomplete data, and removes noisy data. For example, it imputes missing values ​​and removes inappropriate values. This process results in cleansed data. The input is the regional information data in the database, and the output is the cleansed data.

[1286] Step 3:

[1287] The server uses the cleansed data to transform it into a format suitable for machine learning algorithms and trains an AI model. It uses Python's Scikit-learn to vectorize the data and train the model. The input is the cleansed data, and the output is the trained AI model.

[1288] Step 4:

[1289] Users request information about facilities and stores through a smartphone application. For example, they might type "Tell me your recommended ramen restaurant" as text. The input is the user's request, and the output is the request data.

[1290] Step 5:

[1291] The terminal sends the user's request to the server. The server parses the request and searches the database for relevant information. For example, it might perform a search based on the keyword "ramen shop." The input is the request data, and the output is the relevant information.

[1292] Step 6:

[1293] The server generates optimal recommendations using an AI model based on the searched information. It uses a trained model to select highly-rated stores. The input is relevant information, and the output is recommendations.

[1294] Step 7:

[1295] The server sends the generated recommendation information to the user's terminal. For example, it might send a push notification saying, "The ramen shop in front of the station has high ratings." The input is the recommendation information, and the output is the notification sent to the user's terminal.

[1296] Step 8:

[1297] Users visit recommended facilities or shops and then provide feedback. They submit comments such as "It was delicious." The input is the user's feedback, and the output is feedback data.

[1298] Step 9:

[1299] The device sends user feedback to the server. The server stores the feedback data in a database and uses it for retraining. The collected feedback data is also integrated with existing data and used to retrain the AI ​​model. The input is the feedback data, and the output is the updated AI model.

[1300] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1301] This invention is a system for collecting local information data from local governments and individual businesses, recognizing user emotions using an emotion engine based on user requests and feedback, and utilizing that information to recommend the most suitable facilities and businesses. This system operates based on the interaction between a server, a terminal, and a user.

[1302] Collection and storage of local data

[1303] The server receives local information data sent from local governments and individual businesses. The received data is appropriately analyzed, categorized into categories such as "facilities," "events," and "restaurants," and stored in a database. This data forms the basis for generating recommendation information in response to user requests.

[1304] Data preprocessing and AI model training

[1305] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is formatted and converted into a format suitable for machine learning algorithms. This formatted data is used to train the AI ​​model. The newly trained AI model is deployed to the production environment in a timely manner. This enables the provision of highly accurate recommendation information to users.

[1306] Processing user requests and sentiment analysis

[1307] Users request information about facilities and stores through LINE or other applications. The device sends this request to the server. The server analyzes the received request and understands its content. Furthermore, it can use an emotion engine to extract emotional information from the user's request. For example, if a user requests "I'm hungry and very irritated," the emotion engine recognizes "irritated."

[1308] Generation and provision of recommendation information

[1309] The server searches the database for information related to the request. Furthermore, using the retrieved sentiment information, an AI model generates recommendations best suited to the user. For example, a user who is "frustrated" might be recommended a restaurant that can provide service quickly. This recommendation information is then delivered to the user via their device.

[1310] As a concrete example, a user requests via the LINE app, "Tell me about a good ramen restaurant quickly." The device sends this request to the server. The server analyzes the request and uses an emotion engine to obtain "urgent" emotion information. The server searches its database for ramen restaurants that can respond quickly and uses an AI model to recommend, "The ramen restaurant 'Mensho' near the station is highly rated for its quick service." The device then provides this information to the user.

[1311] Gathering feedback and relearning

[1312] After a user visits a recommended facility or store, they provide feedback. The device sends this feedback to a server. The server stores the feedback data in a database and uses it for retraining. The sentiment information included in the provided feedback is also analyzed and stored in the database. This new feedback data is integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[1313] As a concrete example, a user might submit feedback about a ramen restaurant they visited, stating, "It was very delicious, but the service was a little slow." This feedback is sent to the server and stored in the database. The server uses this feedback data and the sentiment information analyzed by the sentiment engine to retrain the AI ​​model and incorporate it into future recommendations.

[1314] Thus, the present invention is a system that efficiently collects local information from local governments and individual shops, analyzes users' emotions using an emotion engine, and provides users with highly accurate recommendation information, thereby stimulating local consumer activity and improving their dwell time.

[1315] The following describes the processing flow.

[1316] Collection and storage of local data

[1317] Step 1:

[1318] The server receives HTTP POST requests from local governments and individual businesses.

[1319] Step 2:

[1320] The server parses the received request body and extracts regional information data.

[1321] Step 3:

[1322] The server categorizes the extracted data into categories such as "facilities," "events," and "restaurants."

[1323] Step 4:

[1324] The server stores the classified data in a database.

[1325] Data preprocessing and AI model training

[1326] Step 1:

[1327] The server periodically checks the raw data in the database, corrects incomplete data, and removes noisy data.

[1328] Step 2:

[1329] The server formats the cleansed data into a format suitable for machine learning algorithms.

[1330] Step 3:

[1331] The server uses the formatted data to train the AI ​​model.

[1332] Step 4:

[1333] The server evaluates the model's performance and adjusts parameters as needed.

[1334] Step 5:

[1335] The server uploads the newly trained model to the production deployment environment.

[1336] Processing user requests and sentiment analysis

[1337] Step 1:

[1338] Users request information about facilities and stores through LINE or other apps.

[1339] Step 2:

[1340] The terminal sends the user's request to the server as an HTTP request.

[1341] Step 3:

[1342] The server analyzes the received request to understand its content and the user's intent.

[1343] Step 4:

[1344] The server uses an emotion engine to extract emotional information from user requests.

[1345] Step 5:

[1346] The server searches the database for information related to the request.

[1347] Step 6:

[1348] The server uses an AI model to generate optimal recommendations based on search results and sentiment information.

[1349] Step 7:

[1350] The server formats the generated recommendation information and sends it to the terminal.

[1351] Step 8:

[1352] The device displays recommendation information on the user interface.

[1353] Specific example

[1354] Step 1:

[1355] The user sends a request in the LINE app saying "Please tell me a delicious ramen shop quickly."

[1356] Step 2:

[1357] The terminal sends this request to the server as an HTTP request.

[1358] Step 3:

[1359] The server analyzes the request and understands that the user is "in a hurry."

[1360] Step 4:

[1361] The server uses the emotion engine to extract the "in a hurry" emotion from the text of the request.

[1362] Step 5:

[1363] The server quickly searches the database for information on ramen shops that can provide services.

[1364] Step 6:

[1365] Based on the search results and emotion information, the server generates optimal recommendation information using an AI model.

[1366] Step 7:

[1367] The server formalizes the recommendation information "The ramen shop 'Noodle Master' near the station is evaluated for quick service" and sends it to the terminal.

[1368] Step 8:

[1369] The terminal displays this information to the user in the LINE app.

[1370] Collection of feedback and re - learning

[1371] Step 1:

[1372] Users provide feedback after using the recommended facilities or stores.

[1373] Step 2:

[1374] The device sends feedback to the server.

[1375] Step 3:

[1376] The server analyzes the feedback and extracts emotional information.

[1377] Step 4:

[1378] The server stores feedback data and emotional information in a database.

[1379] Step 5:

[1380] The server retrains the AI ​​model using the stored feedback data.

[1381] In this way, the system achieves highly accurate information provision and recommendations tailored to user sentiment through an efficient process of collecting, storing, analyzing, learning, recommending, and providing feedback on local information.

[1382] (Example 2)

[1383] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1384] Current regional information recommendation systems have a problem in that they struggle to make recommendations that take into account user emotions and urgency. Furthermore, the presence of data incompleteness and noise often leads to decreased recommendation accuracy. Additionally, there is a lack of effective means to utilize user feedback for retraining, which hinders the continuous improvement of system performance.

[1385] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1386] In this invention, the server includes means for receiving local information data from local governments and individual stores; means for classifying the received local information data and storing it in a database; means for checking the integrity of the stored local information data and correcting and deleting incomplete or noisy data; means for converting the formatted data into a format suitable for machine learning algorithms and training an artificial intelligence model; means for receiving requests from users; means for analyzing the content of requests and obtaining sentiment information using a sentiment engine; means for searching for relevant information from the database and generating recommendation information using an artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; means for receiving feedback from users, storing the received feedback in a database and using it for retraining; and means for optimizing the recommendation information based on the generated sentiment information. This makes it possible to provide highly accurate recommendation information based on the user's sentiment and urgency.

[1387] A "local government" refers to a public institution responsible for managing and operating a specific region.

[1388] "Individual store" refers to a commercial or service facility operated by an individual or a small business.

[1389] "Local information data" refers to data that includes information about facilities, events, restaurants, etc., related to a specific region.

[1390] A "database" refers to a system for systematically storing and managing collected information.

[1391] "Incomplete data" refers to data that is missing necessary information or contains errors.

[1392] "Noise data" refers to unwanted data that hinders analysis and learning.

[1393] A "machine learning algorithm" refers to a method or model for automatically learning patterns and rules from data.

[1394] An "artificial intelligence model" refers to a program that has been trained using machine learning algorithms and possesses the ability to perform specific tasks.

[1395] An "emotion engine" refers to a technology that uses natural language processing and text mining to recognize and analyze emotions from text.

[1396] A "user terminal" refers to an electronic device (e.g., smartphone, tablet, personal computer) used by a user to input and receive information.

[1397] A "request" refers to a request made by a user to a system for the provision of information.

[1398] "Feedback" refers to the evaluations and opinions that users give regarding the information and services provided.

[1399] "Retraining" refers to the process of retraining an artificial intelligence model using new data and feedback.

[1400] "Optimization" refers to adjusting the parameters and processes of a system to their optimal state in order to achieve a specific objective.

[1401] This invention relates to a system that collects local information data from local governments and individual businesses and generates recommendation information based on user requests and feedback. This system operates based on the interaction between a server, a terminal, and a user.

[1402] Collection and storage of local data

[1403] The server receives local information data sent from local governments and individual businesses. This information is retrieved through interfaces such as APIs. Examples include event information from local governments and business hours from individual businesses. This data is categorized into categories such as "facilities," "events," and "restaurants," and stored in a database. Relational databases such as MySQL and PostgreSQL can be used as the database. This data later forms the basis for generating recommendation information in response to user requests.

[1404] Data preprocessing and AI model training

[1405] The server periodically checks the received regional data, correcting and removing incomplete and noisy data. Specifically, it fills in missing information and removes outliers. Data cleansing scripts using Python or Pandas are used. The formatted data is converted into a format suitable for machine learning algorithms. For example, categorical data is encoded into numerical data. This is then used to train the AI ​​model. TensorFlow and PyTorch are used as machine learning frameworks. The newly trained AI model is deployed to the production environment as needed.

[1406] Processing user requests and sentiment analysis

[1407] Users request information about facilities and stores through LINE or a dedicated application. For example, they might send a message like, "Tell me about a good ramen restaurant quickly." The device sends this request to the server. The server analyzes the received request and uses natural language processing technology (e.g., spaCy or NLTK) to understand its content. Furthermore, it uses an emotion engine (e.g., Google Cloud Natural Language API) to extract emotional information from the user's request. For example, it recognizes "irritated" from a request like, "I'm hungry and very irritated."

[1408] Generation and provision of recommendation information

[1409] The server searches the database for information related to the request, and the AI ​​model uses the retrieved sentiment information to generate the most suitable recommendations for the user. For example, for a user who is "frustrated," it recommends restaurants that can provide quick service. This recommendation information is provided to the user through their device. As a specific example, if a user requests "Tell me a good ramen shop quickly" using the LINE app, the device sends the request to the server, and the server analyzes the request and sentiment information. It searches the database for ramen shops that can respond quickly and uses the AI ​​model to recommend "Ramen shop 'XX' near the station is highly rated for its quick service." This information is provided to the user through their device.

[1410] Gathering feedback and relearning

[1411] Users provide feedback after using recommended facilities or stores. For example, they might send feedback via the LINE app, such as, "It was delicious, but the service was a little slow." The device sends this feedback to a server. The server stores the received feedback data in a database and uses it for retraining. The collected feedback data undergoes cleansing and formatting steps, is integrated with existing data, and then used to retrain the AI ​​model. This ensures that the system can always provide highly accurate recommendation information.

[1412] Examples of prompt statements

[1413] The following are specific examples of input prompts for a generative AI model.

[1414] 1. "Can you recommend a good cafe nearby?"

[1415] 2. "I'm looking for a restaurant that's safe and comfortable for families with children."

[1416] 3. "What are some recommended events to go to with friends?"

[1417] 4. "Please recommend a good place for a date."

[1418] By utilizing these prompts, the generative AI model can provide optimal recommendation information based on user requests.

[1419] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1420] Step 1: Collecting local data

[1421] The server receives local information data from local governments and individual businesses. The specific operation is as follows: First, input processing is performed to acquire data through an interface such as an API. Next, the received data is temporarily stored in a buffer. Then, it is classified into categories such as "facilities," "events," and "restaurants," and finally stored in the database. In this step, the input is the received local information data, and the output is the classified and stored data.

[1422] Step 2: Data preprocessing

[1423] The server periodically checks the received regional information data, correcting and removing incomplete and noisy data. Specifically, it performs data integrity checks, supplements missing information, and removes outliers. Next, it formats this data into a format suitable for machine learning algorithms. For example, it encodes categorical data into numerical data. In this step, the input is classified data, and the output is formatted data.

[1424] Step 3: Training the AI ​​model

[1425] The server trains an AI model using pre-processed data. The specific operation is as follows: First, the new data is split into a training set and a test set. Next, the AI ​​model is trained using a machine learning framework such as TensorFlow or PyTorch. The trained model is finally deployed to the production environment. The input in this step is a formatted dataset, and the output is the trained AI model.

[1426] Step 4: Processing User Requests

[1427] Users send requests via LINE or a dedicated application. A concrete example of such a request might be, "Tell me a good ramen restaurant quickly." The device sends this request to the server. The server uses natural language processing techniques to analyze the received request and understand its content. In this step, the input is the user request, and the output is the analyzed request content.

[1428] Step 5: Emotion Analysis

[1429] The server analyzes the received request using an emotion engine to obtain the user's emotion information. Specifically, it uses a natural language processing library to extract keywords and emotional expressions from the request and obtains the emotion information through the emotion engine. For example, from the request "Tell me a good ramen restaurant quickly," the emotion information "I'm in a hurry" is extracted. In this step, the input is the analyzed request content, and the output is the obtained emotion information.

[1430] Step 6: Generating recommendation information

[1431] The server searches the database for relevant information and generates optimal recommendations based on the retrieved sentiment information and user requests. Specifically, it uses an AI model to evaluate the request content and sentiment data to generate optimal recommendations. For example, it might recommend stores that can provide quick service to a user who is "in a hurry." The inputs in this step are sentiment information and user requests, and the output is the generated recommendations.

[1432] Step 7: Providing Recommendation Information

[1433] The server provides the generated recommendation information to the user via the terminal. Specifically, the recommendation information is sent to the terminal via an API, and the terminal displays that information to the user. For example, a recommendation such as "The ramen shop 'XX' in front of the station is highly rated for its fast service" is notified to the LINE app. In this step, the input is the generated recommendation information, and the output is the notification to the user.

[1434] Step 8: Gathering Feedback

[1435] Users provide feedback after using a facility or store. For example, they might say, "The food was delicious, but the service was a little slow." The device sends this feedback to the server, which then stores it in a database. In this step, the input is the user feedback, and the output is the stored feedback data.

[1436] Step 9: Retraining the AI ​​model

[1437] The server retrains the AI ​​model using the collected feedback data. Specifically, it cleanses and formats the new feedback data and integrates it with the existing dataset. Then, it retrains the AI ​​model and deploys it to the production environment. In this step, the input is the saved feedback data, and the output is the retrained AI model.

[1438] (Application Example 2)

[1439] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1440] Traditional local information recommendation systems have a problem of not providing sufficient user satisfaction because they simply provide information without considering the user's emotional state. Furthermore, online virtual shopping lacks personalized recommendations based on user emotions and feedback. This results in a uniform shopping experience for all users, making it difficult to recommend optimal products and services that meet individual needs.

[1441] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for receiving local information data from local governments and individual stores; means for classifying the received local information data and storing it in a database; means for training an artificial intelligence model for generating facility and store recommendations based on user requests using the stored local information data; means for searching for relevant information from the database based on user requests and sentiment information and generating recommendations using the artificial intelligence model; means for transmitting the generated recommendation information to the user terminal; means for receiving user feedback, storing the received feedback in a database and using it for retraining; and means for recommending products and services in the virtual store based on the stored local information data and user sentiment information. This makes it possible to provide detailed information recommendations based on user sentiment and feedback, thereby improving user satisfaction and the shopping experience.

[1442] A "local government" is an administrative body that governs a specific region and is responsible for providing public services and promoting the development of the local community.

[1443] A "private store" is a commercial facility owned and operated by an individual, primarily providing goods or services.

[1444] "Local information data" refers to data provided by local governments or individual businesses that includes various types of information about a particular region.

[1445] A "database" is a system for efficiently storing, searching, and managing data.

[1446] "Emotional information" refers to data that indicates a user's emotional state, and is primarily analyzed by an emotion engine.

[1447] A "user terminal" is a device used by a user to interact with the system, and includes smartphones, tablets, and other similar devices.

[1448] An "artificial intelligence model" is an algorithm or system that automatically performs a specific task by analyzing and learning from data.

[1449] "Feedback" refers to data, including user opinions and impressions after using a system, which is used to improve the system.

[1450] A "virtual store" is a commercial facility in an online environment that provides goods and services via the internet.

[1451] "Recommendation information" refers to information about the most suitable facilities and stores, generated based on user requests and sentiments.

[1452] "Retraining" is the process of retraining an existing artificial intelligence model using new data and feedback.

[1453] This invention is a system that collects local information data and generates and provides recommendation information based on user requests and sentiment information. Specific embodiments of this invention are described below.

[1454] 1. Data collection and storage

[1455] The server receives local information data from local governments and individual businesses. This data includes a wide range of information such as facilities, events, and restaurants. The received data is appropriately analyzed, categorized, and then stored in a database.

[1456] 2. Data preprocessing and AI model training

[1457] The server periodically checks the data in the database, correcting incomplete data and removing noisy data. The cleansed data is formatted to a format suitable for machine learning algorithms, and this is used to train AI models. The newly trained AI models are deployed to the production environment, enabling them to provide users with highly accurate recommendations.

[1458] 3. Processing User Requests and Sentiment Analysis

[1459] Users request information through the virtual store's shopping app or messaging application. The device sends this request to the server. The server analyzes the received request and uses an emotion engine to obtain the user's emotional information. For example, if a user requests, "I'm feeling stressed today, so I'm looking for a relaxing cafe. Do you have any recommendations?", the emotion engine recognizes "stress."

[1460] 4. Generation and provision of recommendation information

[1461] The server searches the database for information related to the request, and using the retrieved sentiment information, the AI ​​model generates recommendations best suited to the user. For example, a user feeling "stressed" would be recommended relaxing cafes or shops. This recommendation information is then sent to the user's device.

[1462] 5. Gathering feedback and relearning

[1463] After a user uses a recommended facility or store, they provide feedback. The device sends this feedback to a server. The server stores the feedback data in a database and uses it for retraining. The sentiment information included in the provided feedback is also analyzed and stored in the database. This new feedback data is integrated with existing data and used to retrain the AI ​​model. This continuously improves the system's recommendation accuracy.

[1464] Hardware and software to be used

[1465] Hardware: Servers, smartphones, tablets

[1466] Software: Web frameworks (e.g., Flask, Django), machine learning libraries (e.g., TensorFlow), sentiment analysis engines (e.g., Emotion-Recognition-API)

[1467] Specific example

[1468] When a user requests, "I'm feeling stressed today, so I'm looking for a relaxing cafe. Do you have any recommendations?", the emotion engine recognizes the user's "stress" and recommends a relaxing establishment. Similarly, if a user requests, "Tell me about a good ramen shop quickly," the system can recommend a ramen shop that can provide quick service based on the emotional information that the user is "in a hurry."

[1469] Example of a prompt:

[1470] "I'm feeling stressed today, so I'm looking for a relaxing cafe. Do you have any recommendations?"

[1471] "Please tell me a good ramen shop quickly. Thank you."

[1472] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1473] Step 1:

[1474] The server receives local information data from local governments and individual businesses. This data is often provided in a structured format such as JSON. The received data is saved to a database in real time. The input is local information data sent from local governments and individual businesses, and the output is the saved data.

[1475] Step 2:

[1476] The server classifies the data in the database and stores it by category. For example, it might classify data into categories such as "restaurants," "events," and "facilities." This classification is performed using a data analysis algorithm. The input is the stored regional information data, and the output is the classified data.

[1477] Step 3:

[1478] The server uses the cleansed data to train an artificial intelligence model. Data shaping processes convert the data into a format suitable for machine learning algorithms, and this is then input into the AI ​​model. This dataset is used to train the model and build a highly accurate recommendation engine. The input is the shaped data, and the output is the trained AI model.

[1479] Step 4:

[1480] The user sends a request via a smartphone or tablet. The request content is parsed using natural language processing. The device sends this request to the server, and the prompt text includes the user's sentiment. The input is the user request, and the output is the parsing result.

[1481] Step 5:

[1482] The server analyzes the received request and uses an emotion engine to obtain the user's emotional information. For example, the emotion engine analyzes emotional information such as "I want to relax," "I'm in a hurry," or "I'm stressed." The input is the user's request, and the output is the user's emotional information.

[1483] Step 6:

[1484] The server searches the database for relevant information and uses an AI model to generate optimal recommendations based on the retrieved sentiment information. For example, based on sentiment information such as "I want to relax," it might recommend quiet cafes or parks. The input is the user's sentiment information and request, and the output is the recommendations.

[1485] Step 7:

[1486] The server sends the generated recommendation information to the user's terminal. The user can receive this information on their smartphone or tablet. The input is recommendation information, and the output is visualized recommendation information.

[1487] Step 8:

[1488] Users utilize a facility or store and then provide feedback. The terminal sends this feedback to a server. The input is the user's feedback, and the output is the analyzed feedback information.

[1489] Step 9:

[1490] The server stores the feedback it receives in a database and uses it for retraining. This allows the AI ​​model to continuously learn and improve its recommendation accuracy. The input is the analyzed feedback information, and the output is the updated AI model.

[1491] The above outlines the specific processing steps of the system that implements the application example.

[1492] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1493] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1494] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1495] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1496] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1497] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1498] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1499] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1500] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1501] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1502] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1503] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1504] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1505] 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.

[1506] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1507] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1508] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1509] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1510] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1511] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1512] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[1513] The following is further disclosed regarding the embodiments described above.

[1514] (Claim 1)

[1515] A means of receiving local information data from local governments and individual businesses,

[1516] A means of classifying received regional information data and storing it in a database,

[1517] A means for training an artificial intelligence model to generate recommendations for facilities and stores based on user requests using stored regional information data,

[1518] A means of receiving requests from users,

[1519] A means for searching a database for relevant information based on a user request and generating recommendations using an artificial intelligence model,

[1520] A means for sending the generated recommendation information to the user's terminal,

[1521] A means of receiving user feedback, storing the received feedback in a database, and using it for retraining,

[1522] A system that includes...

[1523] (Claim 2)

[1524] The system according to claim 1, further comprising means for periodically updating an artificial intelligence model and deploying the updated model to a production environment.

[1525] (Claim 3)

[1526] The system according to claim 1, wherein the user interface for providing recommendation information to the user utilizes a messaging application.

[1527] "Example 1"

[1528] (Claim 1)

[1529] A means of receiving local information data from local governments and individual businesses,

[1530] A method for analyzing received local information data, classifying it into categories such as "facilities," "events," and "restaurants," and storing it in a database,

[1531] A means for periodically cleansing stored regional information data, correcting defects and removing noise data,

[1532] A method for training an artificial intelligence model using a machine learning algorithm with cleansed data,

[1533] Means for deploying trained artificial intelligence models,

[1534] A means of analyzing user requests using natural language processing technology,

[1535] A means for searching for relevant information from a database based on an analyzed request and generating recommendation information using an artificial intelligence model,

[1536] A means for sending the generated recommendation information to the user's terminal,

[1537] A means of receiving user feedback, storing the received feedback in a database, and using it for retraining,

[1538] A system that includes this.

[1539] (Claim 2)

[1540] The system according to claim 1, further comprising means for periodically updating an artificial intelligence model and deploying the updated model to a production environment.

[1541] (Claim 3)

[1542] The system according to claim 1, including a user interface for users to request and receive recommendation information using LINE or a dedicated application.

[1543] "Application Example 1"

[1544] (Claim 1)

[1545] A means of receiving local information data from local governments and individual businesses,

[1546] A means of classifying received regional information data and storing it in a database,

[1547] A means for training an artificial intelligence model to generate recommendations for facilities and stores based on user requests using stored regional information data,

[1548] A means of receiving requests from users,

[1549] A means for searching a database for relevant information based on a user request and generating recommendations using an artificial intelligence model,

[1550] A means for sending the generated recommendation information to the user's terminal,

[1551] A means of receiving user feedback, storing the received feedback in a database, and using it for retraining,

[1552] A means of processing user requests via an application installed on a smartphone,

[1553] A means of providing information on recommended stores and real-time event notifications via push notifications,

[1554] A system that includes this.

[1555] (Claim 2)

[1556] The system according to claim 1, further comprising means for periodically updating an artificial intelligence model and deploying the updated model to a production environment.

[1557] (Claim 3)

[1558] The system according to claim 1, wherein the user interface for providing recommendation information to users utilizes a messaging application and a push notification function.

[1559] "Example 2 of combining an emotion engine"

[1560] (Claim 1)

[1561] A means of receiving local information data from local governments and individual businesses,

[1562] A means of classifying received regional information data and storing it in a database,

[1563] Means for checking the integrity of stored regional information data and correcting and deleting incomplete or noisy data,

[1564] A method for using formatted data to transform it into a format suitable for machine learning algorithms and train artificial intelligence models,

[1565] A means of receiving requests from users,

[1566] A means of analyzing the content of a request and obtaining sentiment information using a sentiment engine,

[1567] A means for searching for relevant information from a database and generating recommendation information using an artificial intelligence model,

[1568] A means for sending the generated recommendation information to the user's terminal,

[1569] A means of receiving user feedback, storing the received feedback in a database, and using it for retraining,

[1570] A means of optimizing recommendation information based on generated sentiment information,

[1571] ...

[1572] A system that includes this.

[1573] (Claim 2)

[1574] The system according to claim 1, further comprising means for periodically updating an artificial intelligence model and deploying the updated model to a production environment.

[1575] (Claim 3)

[1576] The system according to claim 1, wherein the user interface for providing recommendation information to the user utilizes a messaging application.

[1577] "Application example 2 when combining with an emotional engine"

[1578] (Claim 1)

[1579] A means of receiving local information data from local governments and individual businesses,

[1580] A means of classifying received regional information data and storing it in a database,

[1581] A means for training an artificial intelligence model to generate recommendations for facilities and stores based on user requests using stored regional information data,

[1582] A means of receiving requests from users,

[1583] A means for retrieving relevant information from a database based on user requests and sentiment information, and generating recommendations using an artificial intelligence model,

[1584] A means for sending the generated recommendation information to the user's terminal,

[1585] A means of receiving user feedback, storing the received feedback in a database, and using it for retraining,

[1586] A method for recommending products and services within a virtual store based on stored regional information data and user sentiment information,

[1587] A system that includes this.

[1588] (Claim 2)

[1589] The system according to claim 1, further comprising means for periodically updating an artificial intelligence model and deploying the updated model to a production environment.

[1590] (Claim 3)

[1591] The system according to claim 1, wherein the user interface for providing recommendation information to the user utilizes a messaging application and an interactive virtual shopping application. [Explanation of symbols]

[1592] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of receiving local information data from local governments and individual businesses, A means of classifying received regional information data and storing it in a database, A means for training an artificial intelligence model to generate recommendations for facilities and stores based on user requests using stored regional information data, A means of receiving requests from users, A means for searching a database for relevant information based on a user request and generating recommendations using an artificial intelligence model, A means for sending the generated recommendation information to the user's terminal, A means of receiving user feedback, storing the received feedback in a database, and using it for retraining, A system that includes...

2. The system according to claim 1, further comprising means for periodically updating an artificial intelligence model and deploying the updated model to a production environment.

3. The system according to claim 1, wherein the user interface for providing recommendation information to the user utilizes a messaging application.

Images