Multi-role assistant system and device based on large language model in text travel scene

Abstract

The invention discloses a multi-role assistant system and device based on a large language model in a text travel scene. The tourist assistant can understand the natural language problem proposed by the user and provide scenic spot information, a planned tour route, traffic information and peripheral service information; the merchant assistant has the functions of intelligent customer service, operation analysis and strategy suggestion and user evaluation analysis; the government background management assistant can generate a tourism revenue report, and assists market supervision and emergency command. Through large language model training optimization and auxiliary data processing, the tourist experience and merchant benefits are improved, the government is assisted in scientific decision making and supervision, and the development of the travel industry is promoted.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to a multi-role assistant system and device based on a large language model in a cultural and tourism scenario. Background Art

[0002] In the current cultural and tourism scenario, tourists, merchants and government back-end management each face a series of significant problems.

[0003] Tourists often face the problem of too much information and inaccurate information when planning their trip. They need to collect information about the destination's attractions, food, accommodation, transportation, etc. from many channels, but this information is often scattered, incomplete and sometimes contradictory. It may also be difficult to obtain real-time and accurate local information during the trip. In addition, personalized needs are difficult to meet, and it is impossible to customize the most suitable travel plan according to their own interests and time.

[0004] Merchants are faced with the challenges of fierce competition, rising operating costs, and diversified customer needs. In terms of customer service, manual customer service is unable to cope with the number of inquiries during peak periods, resulting in low service efficiency and reduced customer satisfaction. In terms of business decision-making, the lack of accurate data analysis and market trend forecasting makes it difficult to grasp market dynamics and make reasonable adjustments to business strategies. At the same time, it is difficult for merchants to accurately analyze the business direction that needs to be improved from numerous user reviews, and they cannot effectively improve the quality of their business.

[0005] The government backend management is faced with the challenge of effective supervision and scientific planning of the cultural tourism market. In terms of data collection, there are problems with scattered data sources and uneven data quality, making it difficult to fully and accurately grasp the actual situation of the cultural tourism market. In addition, the monitoring and response capabilities for emergencies and market anomalies need to be improved. Summary of the invention

[0006] The purpose of the present invention is to address the deficiencies of the existing technology and propose a multi-role assistant system and device based on a large language model in a cultural and tourism scenario to meet the different needs of different subjects such as tourists, merchants, and government back-end management in cultural and tourism scenarios.

[0007] The object of the present invention is to achieve the following technical solution: a multi-role assistant system based on a large language model in a cultural and tourism scenario, the system comprising:

[0008] The tourist assistant module builds a tourism database containing tourist attraction information. According to the travel needs of tourists, it generates travel route suggestions based on the information in the tourism database through a large language model, generates traffic suggestions based on the information on the traffic data platform, and screens the business information of required merchants based on the merchant database;

[0009] The merchant assistant module builds a merchant database containing merchant business information, screens qualified merchants based on the merchant database through a large language model according to tourists' questions, and feeds back their business information to tourists; and extracts key booking information from user questions according to tourists' booking needs, sends it to the online booking platform, and makes booking operations for tourists according to booking rules and priorities; conducts in-depth mining and analysis of merchant sales information through a large language model, calculates business indicators, and performs sentiment analysis and keyword extraction on the text data of tourists' evaluations, and classifies and summarizes whether tourists are satisfied, as well as their opinions and suggestions;

[0010] The government backend management assistant module collects transaction and revenue data of scenic spot merchants and user evaluation data, analyzes the trend of tourism revenue through a large language model, and extracts key information by combining text data and structured data to evaluate the business behavior of merchants.

[0011] Furthermore, the tourist database in the tourist assistant module periodically collects and updates detailed information of tourist destinations from public databases and online resources, including scenic spot information, opening hours, distances between scenic spots, and transportation methods.

[0012] Furthermore, the large language model in the tourist assistant module is connected and interacted with the real-time traffic data platform to obtain traffic information of scenic spots, including the operating hours, routes, fares and real-time road conditions of transportation tools, as well as the operating hours and fares of sightseeing buses and cable cars in the scenic area; route planning, traffic information and merchant information are generated by the large language model in the form of text descriptions, charts and map annotations, and multiple rounds of dialogue are conducted according to user needs to modify the information.

[0013] Furthermore, the merchant business information includes the merchant's business hours, business scope, product information and preferential policy related information.

[0014] Furthermore, the merchant assistant module builds and maintains a question bank of tourists, classifies and counts tourists' frequently asked questions and the answers of the large language model, and adds them to the list of frequently asked questions.

[0015] Furthermore, through the API interface of the online booking platform service, the merchant assistant module can interact with the booking platform and complete the booking process based on the extracted booking information; and use the payment system of the payment platform service to generate a payment link or QR code based on the booking or purchase information and feedback it to the tourists. After the payment operation is completed, the tourists are provided with instant feedback on the success or failure of the booking.

[0016] Furthermore, the merchant assistant module builds a database of the merchant's sales, inventory and financial information. When the large language model completes a reservation or purchase request initiated by a user, it can automatically record the transaction data into the database. The merchant's own inventory and financial change information is also stored in the database through manual entry or large language model-assisted entry. When the merchant needs to analyze its business conditions, the large language model reads this information, calculates key indicators such as sales, profit margins, and inventory turnover rates, analyzes business trends and potential problems by comparing with historical and industry data, and generates a business conditions analysis report, which is fed back to the merchant through the large language model.

[0017] Furthermore, the large language model in the government backend management assistant module generates detailed monthly, quarterly and annual reports, and presents them in the form of a combination of charts and text.

[0018] Furthermore, the government backend management assistant module interacts with the meteorological, traffic and public safety monitoring systems in real time, uses a large language model to conduct real-time analysis of meteorological and traffic information, quickly identifies key issues and needs, and combines manually set emergency plan rules to generate emergency command strategies and action plans.

[0019] On the other hand, the present invention also provides a multi-role assistant device based on a large language model in a cultural and tourism scenario, including a memory and one or more processors, wherein the memory stores executable code, and when the processor executes the executable code, a multi-role assistant system based on a large language model in a cultural and tourism scenario is implemented.

[0020] Beneficial effects of the present invention:

[0021] The present invention effectively solves the following problems existing in the background technology through a multi-role assistant system based on a large language model in a cultural and tourism scenario:

[0022] For tourists, the problem of large amount of inaccurate information is solved. Through the personalized services of the tourist assistant module, tourists can easily plan travel routes, obtain real-time traffic information and catering and accommodation recommendations, greatly improving the travel experience.

[0023] For merchants, it solves the challenges of rising operating costs and diversified customer demands. The merchant assistant module improves service efficiency through intelligent customer service and automatic reservation services. It helps merchants make reasonable adjustments to their operating strategies through operating status analysis and user evaluation analysis, effectively improving operating quality.

[0024] For the government backend management, it solves the problem of effective supervision and scientific planning of the cultural and tourism market. The government backend management assistant module improves the monitoring and response capabilities of the cultural and tourism market through functions such as tourism revenue reporting, industry supervision and emergency command, and achieves a comprehensive and accurate grasp of the market.

[0025] To sum up, the present invention provides all-round intelligent services and support for different subjects in cultural and tourism scenarios through large language model technology, effectively solving the problems of information asymmetry, low service efficiency and market supervision difficulties existing in the background technology, and has significant practical value and social benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0027] Figure 1 This is a schematic diagram of a multi-role assistant structure based on a large language model in a cultural and tourism scenario provided by the present invention;

[0028] Figure 2 It is a structural diagram of a multi-role assistant device based on a large language model in a cultural and tourism scenario of the present invention. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with the accompanying drawings and implementation examples. It should be understood that the specific implementation examples described herein are only used to explain the present invention and are not used to limit the present invention.

[0030] like Figure 1 As shown, the present invention provides a multi-role assistant system based on a large language model in a cultural and tourism scenario, including a tourist assistant module, a merchant assistant module and a government backend management assistant module.

[0031] Tourist Assistant Module: The Tourist Assistant relies on large language model technology to provide tourists with all-round and personalized services. Through interaction with the Tourist Assistant, tourists can not only easily plan travel routes and obtain scenic spot traffic information, but also obtain key service information such as catering, hotels, and accommodation from the merchant database, greatly improving the travel experience. The functions implemented by the Tourist Assistant module are as follows:

[0032] 1. Build a tourism database to serve the tourist assistant, which collects detailed information about tourist destinations from public databases, online resources and other channels, including attractions, opening hours, distances between attractions, and transportation methods. The database is updated periodically to ensure the accuracy and timeliness of the information.

[0033] 2. Train a large language model that enables the tourist assistant to interact directly with tourists. The large language model is trained using detailed descriptions of tourist attractions provided by the official website of tourist attractions and tourist information in the tourism database as additional training data, so that it can better understand the characteristics of different attractions and the relationship between them and analyze tourism-related information. The trained model is then integrated into the tourist assistant module to ensure that the model can receive user input and provide intelligent responses.

[0034] 3. Design a user interface that allows tourists to input travel requirements, such as travel time, preferred types of attractions (such as natural scenery, history and culture, entertainment and leisure, etc.), budget range, etc. Input these tourist demand information into the big language model, use the big language model to perform data analysis, and comprehensively consider multiple factors such as the popularity of attractions, tourists' interest preferences, travel time constraints, and transportation convenience to generate one or more personalized travel route suggestions, and show tourists a detailed itinerary, including specific attractions to visit, the estimated time to stay at each attraction, and the best transportation methods and routes between attractions. Build a plug-in based on the big language model, which connects to the real-time traffic data platform to obtain the latest traffic information. The technical implementation process of the plug-in first involves creating an interface that can communicate with the real-time traffic data platform API, which is responsible for sending requests and receiving traffic data. The plug-in then parses the data and converts it into a format suitable for algorithm processing. Next, the plug-in inputs the user's itinerary information and real-time traffic data into the algorithm model, waiting for the algorithm model to calculate the recommended traffic plan. After the calculation is completed, the plug-in is responsible for converting the output of the algorithm into a user-friendly format, such as map markers, text descriptions, or charts, and presents it to the user through the user interface. In addition, the plug-in needs to have an error handling mechanism to deal with abnormal situations such as API call failure or data parsing errors to ensure the stability and reliability of the plug-in.

[0035] 4. Based on the previously listed itinerary and current real-time traffic information, provide the fastest and most economical transportation suggestions. The suggestion implementation process includes cleaning and formatting the raw data obtained from the real-time traffic data platform, and then analyzing and extracting key traffic information, such as the operating hours, routes, fares, and real-time traffic information of transportation tools (buses, subways, taxis, and online ride-hailing cars) through predefined algorithm models. When tourists ask about the transportation methods of scenic spots, such as whether it is faster to take the subway or take a taxi from the hotel to a certain scenic spot, the large language model can use the shortest path algorithm to calculate the optimal transportation route based on the user's specific itinerary, and consider multiple factors, such as the speed, cost, and time preference of different transportation tools. These factors are integrated through a weighted scoring system to provide tourists with accurate and timely transportation suggestions, and finally output one or more recommended transportation plans. At the same time, it can also provide tourists with information about transportation within the scenic spot, such as whether there are sightseeing buses and cable cars, their operating hours and fares, etc. In order to cope with real-time changing traffic conditions, the recommended plan is dynamically adjusted according to the specific traffic conditions to ensure that the suggestions provided are always the fastest and most economical.

[0036] 5. Build a plug-in based on the big language model, connect with the merchant database, and obtain detailed information on restaurants and hotels. When tourists need to book restaurants and accommodation, the big language model is used to search and match merchant information according to tourists' needs, and the merchant information that meets tourists' needs is displayed in a structured format, including key information such as name, address, contact information, user evaluation, price range, etc., to help tourists make more appropriate choices. The plug-in first establishes a communication interface connecting the big language model and the merchant database. This interface is implemented using a specific database connection library and the language model API. The interface receives query requests from the big language model and converts them into query statements that the merchant database can understand. When the query statement is sent to the merchant database, the database returns the corresponding result set. The plug-in then processes this result set and extracts merchant information that meets tourists' needs, including key content such as name, address, and contact information. The plug-in uses data structures such as dictionaries or objects to store the information of each merchant, and then passes these structured data to the big language model. The big language model can generate natural language descriptions based on these structured data and display them to tourists. The plug-in also takes error handling into consideration. For example, if the database connection fails, the query statement is wrong, or no qualified merchants are found, corresponding error prompts need to be given to ensure the stability and reliability of the plug-in. The plug-in uses indexes and cache query results during the query process to reduce query time and resource consumption.

[0037] 6. Multimodal data and multi-round dialogue functions: In order to ensure that tourists can clearly understand and effectively use the information provided by the large language model, the large language model will also present complex route planning, traffic information and merchant service information in a concise and intuitive way. The large language model can generate outputs in various forms such as text descriptions, charts, map annotations, etc. to help tourists understand various information more intuitively. In addition, the tourist assistant also supports multi-round dialogues. Tourists can provide further details or modification requirements based on the preliminary information. The large language model will continuously optimize and improve the services provided to meet the ever-changing and refined needs of tourists.

[0038] Merchant Assistant Module: The Merchant Assistant provides all-round support for the business management of merchants through intelligent customer service, automatic reservation service, business status analysis, user evaluation analysis and other functions; specifically, the functions include the following:

[0039] 1. Build a merchant information database to serve the merchant assistant. The database establishes a series of data tables for each merchant, including relevant information such as business hours, business scope, product information, preferential policies, etc. Design a merchant review process to regularly check the information in the data tables of all merchants to ensure the accuracy and timeliness of the database information. Build a plug-in based on the large language model so that the large language model can access the relevant information stored in the database. Implement an intelligent customer service. When a tourist initiates a consultation with the merchant assistant, the intelligent customer service can filter out merchants that meet the user's conditions from the database, and integrate the relevant information stored in the database of these merchants to feedback to the tourist. At the same time, the intelligent customer service will provide a function to browse frequently asked questions. This function will classify and count the frequently asked questions that tourists may ask and the answers of the large language model, and add them to the list of frequently asked questions. In order to ensure the accuracy and professionalism of the answers, the merchant will regularly review and update the database to supplement the latest product information and relevant policies.

[0040] 2. Build and maintain a FAQ database, which collects conversations between tourists and merchant assistants, and selects the most frequent question types or question examples from them through a large language model and inputs them into the database. The database is regularly updated to include new FAQs and answers.

[0041] 3. When a user initiates a reservation or purchase request to the merchant assistant, use the large language model to extract key reservation or purchase information from the user request, and confirm with the user if necessary whether the extracted information is accurate, such as the specified dining time, number of people, special dining requirements, or specific accommodation dates, room preferences, etc., and send this information to the online reservation platform for search and matching, and make reservations for tourists according to the reservation rules and priorities. Implement an API interface that can use the services of the online reservation platform, so that the merchant assistant module can interact with the reservation platform to obtain real-time reservation information, including the number of available dining seats, room types and quantities, reservation prices, etc., and complete the reservation process based on the extracted reservation information. Implement a payment system that can use the services of mainstream payment platforms. After the tourist completes the reservation operation, generate a payment link or QR code based on the reservation or purchase information, and feedback it to the user, simplify the payment process, help tourists realize convenient reservation fee payment and settlement, and provide tourists with instant feedback on the success or failure of the reservation after the payment operation is completed. If there is a reservation failure or reservation conflict, timely feedback to the tourist on relevant information and provide alternative solutions when necessary.

[0042] 4. Build a database that records the merchant's sales, inventory, financial and other information, and build a plug-in based on the big language model to access the database to obtain sales data, inventory levels, cost expenditures and other information. When the big language model completes the reservation or purchase request initiated by the user, it can automatically record the transaction data into the database. The merchant's own inventory and financial change information can also be stored in the database through manual entry or large language model-assisted entry. When the merchant needs to analyze the business situation, the big language model reads this information, calculates key indicators such as sales, profit margins, and inventory turnover rates, and analyzes business trends and potential problems by comparing with historical and industry data, and generates a business situation analysis report, which is fed back to the merchant through the big language model.

[0043] 5. Collect user evaluation data from multiple channels by visiting various websites, including online reviews, social media, and the merchant's own evaluation system. Input the collected user evaluation data into the large language model for classification, keyword extraction, and sentiment analysis to determine user satisfaction and the user's main concerns. The large language model will further classify and summarize the user's specific opinions and suggestions, such as evaluations on the taste of dishes, service quality, environmental comfort, price rationality, etc., and provide the analysis results and suggestions to the merchant. Through comprehensive analysis of a large amount of evaluation data, it provides the merchant with directions for improvement and optimization.

[0044] Government backend management assistant module: The government backend management assistant module provides the government with a variety of key functions through a large language model, including tourism revenue reporting, industry supervision, emergency command, etc.; specifically, the following functions are included:

[0045] 1. Integrate transaction and revenue data interfaces from various merchants (such as scenic spots, hotels, etc.) to automatically collect revenue data from different sources such as scenic spots, hotels, restaurants and travel agencies, and use big data analysis technology to clean the data, remove invalid or erroneous data, and classify and summarize the data. Input the collected valid data into the big language model, use the big language model to build a trend prediction model, analyze the trend of tourism revenue, generate detailed monthly, quarterly and annual reports, and present the analysis results to government management departments in the form of charts and text, so that the government can intuitively understand the overall situation and growth trend of tourism revenue and provide data support for decision-making.

[0046] 2. Collect user evaluation data from multiple channels by visiting various websites, input these evaluation information and the above-collected transaction and revenue data into the big language model, and use the big language model to extract key information from these text data and structured data, so as to evaluate the business behavior of merchants and identify irregular behaviors and risk points. Compared with manual processing of these industry regulatory data, the efficiency and effectiveness of industry supervision are improved. If the big language model analyzes that the merchant has irregular behavior, it will prompt the government management department to conduct manual review and confirmation.

[0047] 3. Through the establishment of real-time data interaction interfaces with meteorological, transportation, public safety and other monitoring systems, the large language model is used to analyze meteorological, transportation, public safety and other information in real time to quickly identify key issues and needs. If there is an emergency, based on the analysis results and model, the large language model is used in combination with the manually set emergency plan rules to generate emergency command strategies and action plans, providing the government with support for rapid response and effective disposal. This function can coordinate relevant departments and resources to achieve rapid response and effective disposal of emergencies and ensure the safety of tourists and the public.

[0048] Corresponding to the aforementioned embodiment of a multi-role assistant system based on a large language model in a cultural and tourism scenario, the present invention also provides an embodiment of a multi-role assistant device based on a large language model in a cultural and tourism scenario.

[0049] See also Figure 2 A multi-role assistant device based on a large language model in a cultural and tourism scenario provided in an embodiment of the present invention includes a memory and one or more processors, wherein the memory stores executable code, and when the processor executes the executable code, it is used to implement a multi-role assistant system based on a large language model in a cultural and tourism scenario in the above embodiment.

[0050] The embodiment of a multi-role assistant device based on a large language model in a cultural and tourism scenario provided by the present invention can be applied to any device with data processing capabilities, and the any device with data processing capabilities can be a device or apparatus such as a computer. The device embodiment can be implemented through software, or through hardware or a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, it is formed by the processor of any device with data processing capabilities in which it is located reading the corresponding computer program instructions in the non-volatile memory into the internal memory for execution. From a hardware perspective, if Figure 2 As shown, it is a hardware structure diagram of any device with data processing capability in which a multi-role assistant device based on a large language model in a cultural tourism scene provided by the present invention is located. Figure 2 In addition to the processor, memory, network interface, and non-volatile memory shown, any device with data processing capabilities in which the apparatus in the embodiments is located may also include other hardware, generally based on the actual functions of the device with data processing capabilities, which will not be described in detail.

[0051] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0052] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of the present invention. Ordinary technicians in this field can understand and implement it without paying creative work.

[0053] An embodiment of the present invention also provides a computer-readable storage medium on which a program is stored. When the program is executed by a processor, a multi-role assistant system based on a large language model in a cultural and tourism scenario in the above embodiment is implemented.

[0054] The computer-readable storage medium may be an internal storage unit of any device with data processing capability described in any of the aforementioned embodiments, such as a hard disk or a memory. The computer-readable storage medium may also be an external storage device of any device with data processing capability, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit and an external storage device of any device with data processing capability. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capability, and may also be used to temporarily store data that has been output or is to be output.

[0055] The present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the multi-role assistant system based on a large language model in a cultural and tourism scenario.

[0056] The above embodiments are used to illustrate the present invention rather than to limit the present invention. Any modification and change made to the present invention within the spirit of the present invention and the protection scope of the claims shall fall within the protection scope of the present invention.

Claims

1. A multi-role assistant system based on a large language model in a cultural and tourism scenario, characterized in that: The system includes: The tourist assistant module builds a tourism database containing tourist attraction information. According to the travel needs of tourists, it generates travel route suggestions based on the information in the tourism database through a large language model, generates traffic suggestions based on the information on the traffic data platform, and screens the business information of required merchants based on the merchant database; The merchant assistant module builds a merchant database containing merchant business information, screens qualified merchants based on the merchant database through a large language model according to tourists' questions, and feeds back their business information to tourists; and extracts key booking information from user questions according to tourists' booking needs, sends it to the online booking platform, and makes booking operations for tourists according to booking rules and priorities; conducts in-depth mining and analysis of merchant sales information through a large language model, calculates business indicators, and performs sentiment analysis and keyword extraction on the text data of tourists' evaluations, and classifies and summarizes whether tourists are satisfied, as well as their opinions and suggestions; The government backend management assistant module collects transaction and revenue data of scenic spot merchants and user evaluation data, analyzes the trend of tourism revenue through a large language model, and extracts key information by combining text data and structured data to evaluate the business behavior of merchants.

2. According to claim 1, a multi-role assistant system based on a large language model in a cultural and tourism scenario is characterized in that: The tourist database in the tourist assistant module periodically collects and updates detailed information of tourist destinations from public databases and online resources, including scenic spot information, opening hours, distances between scenic spots, and transportation methods.

3. According to claim 1, a multi-role assistant system based on a large language model in a cultural and tourism scenario is characterized in that: The large language model in the tourist assistant module is connected and interacted with the real-time traffic data platform to obtain traffic information of scenic spots, including the operating hours, routes, fares and real-time road conditions of transportation tools, as well as the operating hours and fares of sightseeing buses and cable cars in the scenic area; route planning, traffic information and merchant information are generated by the large language model in the form of text descriptions, charts and map annotations, and multiple rounds of dialogue are conducted according to user needs to modify the information.

4. According to claim 1, a multi-role assistant system based on a large language model in a cultural and tourism scenario is characterized in that: The merchant business information includes the merchant's business hours, business scope, product information and preferential policy related information.

5. According to claim 1, a multi-role assistant system based on a large language model in a cultural and tourism scenario is characterized in that: The merchant assistant module builds and maintains a question bank for tourists, classifies and counts tourists' frequently asked questions and the answers from the large language model, and adds them to the list of frequently asked questions.

6. According to claim 1, a multi-role assistant system based on a large language model in a cultural and tourism scenario is characterized in that: Through the API interface of the online booking platform service, the merchant assistant module can interact with the booking platform and complete the booking process based on the extracted booking information; and use the payment system of the payment platform service to generate a payment link or QR code based on the booking or purchase information and feedback it to the tourists. After the payment operation is completed, the tourists will be provided with instant feedback on the success or failure of the booking.

7. According to claim 1, a multi-role assistant system based on a large language model in a cultural and tourism scenario is characterized in that: The merchant assistant module builds a database of the merchant's sales, inventory and financial information. When the large language model completes the reservation or purchase request initiated by the user, it can automatically record the transaction data into the database. The merchant's own inventory and financial change information is also stored in the database by manual entry or large language model assisted entry; When merchants need to analyze their business conditions, they can use the big language model to read this information, calculate key indicators such as sales, profit margins, and inventory turnover, analyze business trends and potential problems by comparing them with historical and industry data, and generate business status analysis reports, which will be fed back to merchants through the big language model.

8. According to claim 1, a multi-role assistant system based on a large language model in a cultural and tourism scenario is characterized in that: The large language model in the government backend management assistant module generates detailed monthly, quarterly and annual reports, and presents them in the form of a combination of charts and text.

9. According to claim 1, a multi-role assistant system based on a large language model in a cultural and tourism scenario is characterized in that: The government backend management assistant module interacts with meteorological, traffic and public safety monitoring systems in real time, uses a large language model to conduct real-time analysis of meteorological and traffic information, quickly identifies key issues and needs, and generates emergency command strategies and action plans based on manually set emergency plan rules.

10. A multi-role assistant device based on a large language model in a cultural and tourism scenario, comprising a memory and one or more processors, wherein the memory stores executable code, characterized in that: When the processor executes the executable code, a multi-role assistant system based on a large language model in a cultural and tourism scenario is implemented as described in any one of claims 1-9.

Images