Smart tourism service platform and method for city-level tourist attractions

Abstract

The invention discloses a smart tourism service platform and method for city-level tourism spots, and relates to the technical field of smart tourism services. The method comprises the following steps: receiving interest preference and itinerary constraint conditions input by a user, generating a personalized initial itinerary planning scheme through a recommendation algorithm on the basis of a city-level tourism resource database, and responding to a confirmation and modification instruction of the user on the initial itinerary planning scheme; one-stop online reservation and payment of scenic spot tickets, accommodation and traffic services are completed, a unified electronic certificate is generated, and in the touring process of a user, based on geographical location information of a user terminal, multimedia tour guide content of the current or adjacent scenic spots is automatically triggered and pushed, so that the user experience is improved. And based on the real-time position data and behavior data generated in the touring process of the user, the follow-up scenic spot, catering or service recommendation information is dynamically adjusted and pushed, so that the whole-course, coherent and efficient smart tourism experience is provided for the user.

Description

Technical Field

[0001] This invention relates to the field of smart tourism service technology, and in particular to a smart tourism service platform and method for city-level tourist attractions. Background Technology

[0002] With the improvement of the national economic level and the popularization of leisure concepts, tourism has become an important engine for promoting regional economic development. City-level tourism, in particular, attracts a large number of tourists due to its convenient transportation, well-developed supporting facilities, and concentrated cultural resources. Currently, smart tourism services on the market mainly rely on large online travel platforms, providing users with basic functions such as travel information inquiry, flight and hotel booking, and scenic spot ticket purchase through websites and mobile applications. Their core recommendation technologies are mostly based on collaborative filtering algorithms and search popularity ranking, recommending popular travel products and destinations to users by analyzing users' historical behavioral data and group behavior patterns. Although existing technologies have solved the problem of information asymmetry to some extent, the following significant technical shortcomings still exist in providing refined and intelligent services for city-level tourist attractions: 1) Existing platforms generate itineraries based on simple rule combinations or popular attractions, lacking a deep understanding of user preferences. Their recommendation algorithms fail to effectively integrate user profiles, real-time contextual information (such as geographical location, weather, and physical condition), and complex itinerary constraints (such as time, budget, and transportation connections), resulting in itineraries lacking a reasonable spatiotemporal sequence and failing to provide users with a truly customized "one-stop" solution.

[0003] 2) Most existing services stop at the pre-trip booking stage. During the actual visit, the platform lacks effective technical means for real-time sensing and interaction. When unforeseen circumstances arise (such as crowded attractions, sudden weather changes, or changes in user interests), the system cannot provide timely and effective dynamic itinerary adjustment suggestions, leading to interruptions in the visitor experience and even negative emotions.

[0004] 3) City-level tourism resources involve multiple entities such as scenic spots, hotels, transportation, and catering, with inconsistent data standards. Existing platforms struggle to achieve deep integration and standardized processing of cross-domain, multi-source, heterogeneous data, resulting in fragmentation in the service chain across the "planning-booking-navigation-feedback" stages, failing to create a seamless closed-loop service experience.

[0005] In summary, to address the significant shortcomings of traditional tourism service models in terms of informatization and intelligentization, and to meet the growing demand for personalized and in-depth experiences from modern tourists, a smart tourism service platform and methodology for city-level tourist attractions is proposed. Summary of the Invention

[0006] The main objective of this invention is to provide a smart tourism service platform and method for city-level tourist attractions. This platform can deeply integrate multi-source data, utilize interpretable and optimizable intelligent algorithms to generate personalized and rational itineraries for users before their visit, provide context-aware dynamic services during the visit, and achieve self-evolution through feedback after the visit, thereby providing users with a coherent and efficient smart tourism experience throughout the entire trip.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A smart tourism service method for city-level tourist attractions includes the following steps: Step S1: Receive user input of interests and travel constraints, and generate a personalized initial travel itinerary plan based on a city-level tourism resource database using a recommendation algorithm. The plan includes at least one attraction, accommodation, and transportation information. Step S2: In response to the user's confirmation and modification instructions for the initial travel itinerary plan, complete the one-stop online booking and payment for scenic spot tickets, accommodation and transportation services, and generate a unified electronic voucher; Step S3: During the user's tour, based on the user's terminal's geographical location information, automatically trigger and push multimedia guide content for the current or nearby attractions; Step S4: Based on the real-time location and behavior data generated by the user during the tour, dynamically adjust and push subsequent recommendations for attractions, restaurants, or services; Step S5: Receive and publish the evaluation information and travelogue content submitted by users after the tour, and optimize and iterate the city-level tourism resource database and recommendation algorithm based on the evaluation data and behavioral data of all users.

[0008] Furthermore, in step S1, the step of basing the information on the city-level tourism resource database specifically includes: A unified tourism resource database is constructed and maintained. The database integrates data on attractions, hotels, transportation, catering, and activities from multiple sources through a data interface, and performs standardization and consistency processing on the data to support data calls for the recommendation algorithm.

[0009] Furthermore, in step S1, the specific steps of generating a personalized initial travel itinerary plan using a recommendation algorithm include: Based on the user's registration information, historical behavior data, and interests and preferences, a user profile is constructed; The user profile and the itinerary constraints are matched with the tourism resource database, and a recommendation algorithm is used to generate a rational itinerary plan that includes time series and geographical location series. The recommendation algorithm is a mathematical model based on logistic regression, expressed as: ; in, Let represent the probability of choosing the next best destination i at step t of the journey; e is the natural constant. b is the model weight; b is the bias term; The sigmoid function maps the output of a linear combination to the interval (0,1), and is represented as a probability. Let i be the feature vector of the scenic spot i.

[0010] Furthermore, in step S2, the one-stop online booking and payment specifically includes: It connects to multiple third-party ticketing, accommodation and transportation service provider systems through a unified backend service interface; The platform provides users with a unified booking interface and payment gateway at the front end, allowing users to complete the booking and payment of multiple different services in a single session; Step S2 also includes: Generate an electronic voucher with a QR code or barcode corresponding to the reservation order; At the verification point, the order is verified by scanning the electronic voucher, and the verification status is synchronized to the platform database in real time.

[0011] Furthermore, in step S3, the automatic triggering and pushing of multimedia guide content for the current or nearby attractions specifically includes: The user's real-time geographical location is obtained through GPS, base station, or Wi-Fi positioning technology; When a user is determined to have entered a preset electronic fence area, the system automatically pushes voice explanations, graphic introductions, or augmented reality interactive content corresponding to that area to the user's terminal.

[0012] Furthermore, step S4 specifically includes: Real-time analysis of users' movement patterns, dwell time, and consumption behavior; When a user's original itinerary is detected to have changed or that there is a free period, suggestions for alternative attractions, restaurants, or leisure activities are recalculated and pushed based on the user's real-time location, current time, and user profile.

[0013] Furthermore, in step S5, the optimization and iteration of the city-level tourism resource database and recommendation algorithm based on the evaluation data and behavioral data of all users specifically includes: Utilize natural language processing techniques to analyze the sentiment and keywords in user review texts; Sentiment analysis results, trends in tourist attraction popularity, and user behavior preferences are used as feedback data to adjust the information weights in the tourism resource database and optimize the parameters of the recommendation algorithm.

[0014] Furthermore, the feature vector of the scenic spot i At least including: User-item matching features used to quantify the fit between user preferences and attraction attributes. ; Trip constraint features used to ensure the feasibility of a trip in terms of time, budget, and geography. ; Contextual and sequential features used to optimize the experience quality and logical coherence of a trip. ; Among them, the user-item matching feature User profile vectors Image vector of scenic spot i The cosine similarity between them is evaluated and expressed as: ; The travel constraint feature Including the cost of travel from the previous attraction to candidate attraction i Tour time cost Tour cost Geographical distance, time availability and budget availability ; The context and sequence features This includes an indicator variable on whether candidate attraction i is in the same area as the previous attraction, the category diversity entropy of the itinerary after adding candidate attractions, the normalized popularity of candidate attractions, and a fatigue index that represents the user's current level of fatigue. The category diversity entropy cd is calculated using the following formula: A collection of attraction types, This represents the proportion of type c attractions in the itinerary after adding candidate attractions; The fatigue index fl is calculated using the following formula: The preset intensity label value for attraction i. This is the maximum value for the preset intensity label.

[0015] Furthermore, the method provides services to users through a web application or mini-program developed based on the Vue.js front-end framework.

[0016] A smart tourism service platform for city-level tourist attractions, used to implement a smart tourism service method for city-level tourist attractions, including: The itinerary planning module is used to generate personalized initial travel itinerary plans. The reservation and payment module is used to complete one-stop online reservations and payments and generate electronic vouchers; The LBS navigation module is used to trigger and push multimedia navigation content based on the user's location. The dynamic recommendation module is used to dynamically adjust and push recommendation information during the tour; The data collection and feedback module is used to receive user reviews and travelogues and to collect the data. The data management module is used to store, integrate, and standardize tourism resource data, and to perform algorithm optimization iterations.

[0017] The present invention has the following beneficial effects: Compared with existing technologies, this solution introduces a logistic regression model and constructs a multi-dimensional feature vector that integrates user profiles, real-time constraints, and context sequences. This allows for a quantitative balance of multiple factors such as user preferences, time, budget, geographical location, and fatigue, resulting in an optimized route tailored to the current user.

[0018] Compared with existing technologies, this solution provides users with an immersive guided tour experience through LBS electronic fence technology, transforming the service from passive query to proactive push. When unplanned events occur (such as sudden weather changes, attraction closures, or user delays), the iterative decision-making process can be restarted from the current state to generate new and feasible follow-up itineraries, ensuring the continuity and high quality of the travel experience and significantly improving tourists' sense of security and satisfaction.

[0019] Compared with existing technologies, this solution collects post-travel feedback data and uses the cross-entropy loss function to retrain the model, forming a complete closed loop from "prediction-execution-feedback-learning". It can continuously fine-tune as user data accumulates, increasingly understand user needs, and thus improve recommendation accuracy and service quality. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating a smart tourism service method for city-level tourist attractions according to the present invention. Figure 2 This is a schematic diagram of the decision-making process for itinerary planning in the present invention. Figure 3 This is a schematic diagram of the structure of a smart tourism service platform for city-level tourist attractions according to the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0022] See Figure 1The flowchart of the smart tourism service method for city-level tourist attractions of the present invention, shown below, includes the following steps: Step S1: Receive user input of interests and travel constraints, and generate a personalized initial travel itinerary plan based on a city-level tourism resource database using a recommendation algorithm. The plan includes information on at least one attraction, accommodation, and transportation. Step S2: In response to the user's confirmation and modification instructions for the initial travel itinerary plan, complete the one-stop online booking and payment for scenic spot tickets, accommodation and transportation services, and generate a unified electronic voucher; Step S3: During the user's tour, based on the user's terminal's geographical location information, automatically trigger and push multimedia guide content for the current or nearby attractions; Step S4: Based on the real-time location and behavior data generated by the user during the tour, dynamically adjust and push subsequent recommendations for attractions, restaurants, or services; Step S5: Receive and publish the evaluation information and travelogue content submitted by users after the tour, and optimize and iterate the city-level tourism resource database and recommendation algorithm based on the evaluation data and behavioral data of all users.

[0023] The core of this invention lies in decomposing the trip planning problem into a series of sequential decision problems driven by a logistic regression model. For its overall architecture and decision-making process, please refer to [link / reference needed]. Figure 2 As shown, the specific steps include: Step 1: Iterative Trip Planning Based on Logistic Regression Step 1.1: Input Reception and Initialization: The system receives the user's input of trip constraints C (such as total number of days D, total budget B, interest preference tags). It initializes an empty trip and the current state (such as current time, current location, and budget spent).

[0024] Step 1.2: Iterative attraction selection (for each available spot in the itinerary): a) Candidate set generation: Select attractions that meet basic conditions (such as being open and not exceeding the total budget) from the tourism resource database as candidates.

[0025] b) Feature Engineering: For each candidate attraction i, under the current travel state t, calculate its combined feature vector. ,include: User-item matching features used to quantify the fit between user preferences and attraction attributes. ; Trip constraint features used to ensure the feasibility of a trip in terms of time, budget, and geography. ; Contextual and sequential features used to optimize the experience quality and logical coherence of a trip. ; Among them, user-item matching features User profile vectors Image vector of scenic spot i The cosine similarity between them is evaluated and expressed as: ; Stroke constraint features Including the cost of travel from the previous attraction to candidate attraction i Tour time cost Tour cost Geographical distance, time availability and budget availability ; Context and sequence features This includes an indicator variable on whether candidate attraction i is in the same area as the previous attraction, the category diversity entropy of the itinerary after adding candidate attractions, the normalized popularity of candidate attractions, and a fatigue index that represents the user's current level of fatigue. The category diversity entropy (cd) is calculated using the following formula: A collection of attraction types, This represents the proportion of type c attractions in the itinerary after adding candidate attractions; The fatigue index fl is calculated using the following formula: The preset intensity label value for attraction i. This is the maximum value for the preset intensity label.

[0026] c) Probability prediction: Predicting the feature vector Input the pre-trained logistic regression model: This probability represents the suitability of selecting attraction i in the current step.

[0027] d) Decision-making and filtering: First, filter the candidate set based on hard constraints (such as "whether it is within opening hours" or "whether it exceeds the remaining budget for the day"). Then, among the remaining attractions, select the attraction with the highest probability P as the next stop on the itinerary.

[0028] e) Status update: Add the selected attraction i to the itinerary S and update the current status (current location, time and budget spent, fatigue level, etc.).

[0029] Step 1.3: Loop and Termination: Repeat step 1.2 until the itinerary is filled with the reserved number of days or no more new attractions can be added, and finally generate a complete itinerary plan S.

[0030] Step 2: One-stop booking and payment Resource Confirmation: After the user confirms the final itinerary, the system initiates real-time inventory and price inquiries with various third-party service providers (ticketing, hotels, transportation) through a unified backend interface.

[0031] Unified Booking: Users can select all the services they need to book (attraction tickets, hotel accommodations, intercity train tickets, etc.) at once on the platform's unified interface.

[0032] Aggregated payment: The system calls a unified payment gateway, and users complete the payment for all orders using a single payment method (such as WeChat Pay).

[0033] Electronic voucher generation: After successful payment, the system generates an electronic order containing all booking information and generates a unique QR code or barcode for each redeemable item (such as tickets).

[0034] Step 3: LBS-guided tours and real-time interaction Location monitoring: The system continuously monitors the user's geographical location (via the app or mini-program) during the user's travel.

[0035] Content triggering: When the system determines that a user has entered a preset area of ​​a scenic spot through electronic fence technology, it will automatically push multimedia guide content of the scenic spot to the user's terminal, such as audio guide, historical story text and images, 360° panoramic pictures or AR real-view navigation.

[0036] Service integration: It also provides information on nearby emergency assistance, restrooms, restaurants, shopping points, and other practical services.

[0037] Step 4: Dynamic Recommendation and Adjustment Behavioral analysis: The system analyzes user behavior data in real time, such as the attractions actually visited, the length of time spent at each location, movement speed, and consumption records.

[0038] Dynamic Adjustment: Based on real-time analysis results, subsequent recommendations are dynamically adjusted. For example, if a user stays at an attraction for far longer than planned, the system will prompt adjustments to the subsequent itinerary and recommend attractions where time can be omitted or shortened; if it suddenly rains, indoor activities will be prioritized.

[0039] Step 5: Feedback and System Evolution Feedback collection: After the trip, the system will proactively invite users to rate and evaluate the attractions they visited, the hotels they stayed in, and the overall itinerary planning, and encourage users to publish travelogues with pictures and text in the "Travel Notes" community.

[0040] Model optimization: The system uses the user's actual trip data, rating data, and UGC content as new training data, and uses machine learning algorithms (such as logistic regression models) to iteratively update the recommendation model, making its future recommendations more accurate.

[0041] Model retraining: Use high-rated trips actually performed by users as positive samples to construct a training dataset. .

[0042] Parameter Update: Using gradient descent, the weights of the logistic regression model are updated with the objective of minimizing the cross-entropy loss function L. The bias b makes the model's predictions increasingly match the user's true preferences. Specifically, the loss function L is expressed as: .

[0043] The technical solution of the present invention will be further explained below with reference to two specific application scenarios: Application Scenario 1: Planning efficient half-day leisure itineraries for business travelers User background: A company executive who is on a business trip to Changchun has half a day of free time and wants to take a relaxing city tour.

[0044] Logistic Regression Model Decision Process: 1) Initial state: Constraints = {Time: 4 hours, Starting location: Renmin Street, Preferences: Relaxed, Historical}.

[0045] 2) First round of selection: Candidate attractions: Puppet Emperor's Palace Museum, Changchun Film Studio Site Museum, Jingyuetan Forest Park.

[0046] Feature Calculation and Prediction: For "The Puppet Emperor's Palace," its feature vector may reflect: high cosine similarity to the user's preference for "history," low movement cost, moderate tour time, and low fatigue. The model calculates that it has the highest probability.

[0047] Decision: Choose the "Puppet Emperor's Palace of Manchukuo" as the first stop on the itinerary.

[0048] 3) Second round of selection: Current status updated to: Location - Puppet Emperor's Palace, Remaining Time Fatigue level - slightly increased after 24 hours.

[0049] Feature calculation and prediction: For "Changying", its sequence features (both are indoor museums, complementary to but different from "Puppet Manchukuo Palace") and constraint features (close distance, time) make its probability the highest in the second round.

[0050] Decision: Choose Changchun Film Studio as the second stop.

[0051] 4) Output: Generate the final solution [Puppet Emperor's Palace Museum -> Changchun Film Studio Site Museum]. This solution highly meets the user's needs for "efficiency, ease, and historical appeal".

[0052] Application Scenario 2: Dynamically adjusting itineraries for family trips on rainy days User background: A family visiting Changbai Mountain, who originally planned to go hiking in the afternoon.

[0053] Dynamic adjustment process: 1) Triggering event: It started raining heavily at noon, and the system sensed this change through the weather API it accessed.

[0054] 2) Replanning: The system restarts the iterative attraction selection process, starting from the user's current hotel.

[0055] 3) Changes in model decision-making: The contextual features of all outdoor attractions (such as hiking trails) being equal to 0 become a strong negative factor on rainy days, causing the probability calculated by the logistic regression model to drop sharply.

[0056] For indoor attractions (such as volcanic hot spring resorts and folk museums), the context feature = 1 becomes a strong positive factor, and the probability increases significantly.

[0057] Meanwhile, hot springs, with their "relaxing" intensity label, are even more appealing to families who have already visited the morning's attractions due to their fatigue level.

[0058] 4) Result: The system immediately pushed an adjustment suggestion to the user: "Due to weather conditions, we recommend a nearby volcanic hot spring resort as an alternative, which you can book immediately." The family user adopted the suggestion and had a good experience.

[0059] This invention also provides a smart tourism service platform for city-level tourist attractions, used to implement a smart tourism service method for city-level tourist attractions, the structural diagram of which is shown below. Figure 3 As shown, it includes: The itinerary planning module is used to generate personalized initial travel itinerary plans. The reservation and payment module is used to complete one-stop online reservations and payments and generate electronic vouchers; The LBS navigation module is used to trigger and push multimedia navigation content based on the user's location. The dynamic recommendation module is used to dynamically adjust and push recommendation information during the tour; The data collection and feedback module is used to receive user reviews and travelogues and to collect the data. The data management module is used to store, integrate, and standardize tourism resource data, and to perform algorithm optimization iterations.

[0060] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1.A smart tourism service method for urban-level tourist attractions, characterized by, The method comprises the following steps: Step S1: receiving user inputted interest preferences and travel constraints, generating an initial personalized travel itinerary plan based on a city-level tourism resource database and a recommendation algorithm, the plan including at least one scenic spot, accommodation and transportation information; Step S2: in response to user confirmation and modification instructions for the initial travel itinerary plan, completing one-stop online booking and payment of scenic spot tickets, accommodation and transportation services, and generating unified electronic vouchers; Step S3: during user travel, automatically triggering and pushing multimedia guide content of the current or adjacent scenic spot based on the geographic location information of the user terminal; Step S4: based on real-time location data and behavior data generated by the user during the travel process, dynamically adjusting and pushing subsequent scenic spot, catering or service recommendation information; Step S5: receiving and publishing user evaluation information and travelogue content submitted after the end of the travel, and based on all user evaluation data and behavior data, optimizing and iterating the city-level tourism resource database and the recommendation algorithm. 2.The smart tourism service method for urban tourism attractions according to claim 1, wherein, In step S1, the city-level tourism resource database specifically includes: A unified tourism resource database is constructed and maintained, which integrates scenic spot, hotel, transportation, catering and activity data from multiple sources through a data interface, and standardizes and processes the data for consistency to support data calling of the recommendation algorithm. 3.The city-level tourist attraction-oriented smart tourism service method of claim 1, wherein, In step S1, the initial personalized travel itinerary plan generated by the recommendation algorithm specifically includes: Based on user registration information, historical behavior data and interest preferences, a user portrait is constructed; The user portrait, travel constraints and tourism resource database are matched, and a reasonable itinerary plan containing time sequence and geographic location sequence is generated using a recommendation algorithm; The recommendation algorithm is a mathematical model based on logistic regression, represented as: ; where, represents the probability of selecting the next best attraction i at step t of the itinerary; e is the natural constant; is the model weight; b is the bias term; is the sigmoid function that maps the output of the linear combination to the interval (0, 1) and represents the probability; is the feature vector of attraction i. 4.The city-level tourist attraction-oriented smart tourism service method of claim 1, wherein, In step S2, the one-stop online booking and payment specifically includes: Through a unified backend service interface, multiple third-party ticketing, accommodation and transportation service provider systems are connected; A unified booking interface and payment gateway is provided for users on the platform front end, and users complete multiple different service booking and payment operations in one session; The step S2 further includes: Generating a two-dimensional code or bar code electronic voucher corresponding to the booking order; At the check-in point, the order is checked out by scanning the electronic voucher, and the check-out status is synchronized to the platform database in real time. 5.The city-level tourist attraction-oriented smart tourism service method of claim 1, wherein, In step S3, the automatic triggering and pushing of multimedia guide content of the current or adjacent scenic spot specifically includes: The user's real-time geographic location is obtained through GPS, base station or Wi-Fi positioning technology; When the user enters the preset electronic fence area, the corresponding voice commentary, text introduction or augmented reality interactive content of the area is automatically pushed to the user terminal. 6.The city-level tourist attraction-oriented smart tourism service method of claim 1, wherein, The step S4 specifically includes: Real-time analysis of user movement trajectory, stay duration and consumption behavior; When the user's original itinerary is changed or an idle period is detected, based on the user's real-time location, current time and user portrait, alternative scenic spot, restaurant or leisure activity suggestions are recalculated and pushed. 7.The city-level tourist attraction-oriented smart tourism service method of claim 1, wherein, In step S5, the evaluation data and behavior data of all users are used to optimize and iterate the city-level tourism resource database and the recommendation algorithm, specifically including: Using natural language processing technology to analyze the sentiment tendency and keywords of user evaluation text; Using the results of sentiment analysis, the trend of scenic spot popularity and user behavior preferences as feedback data to adjust the information weight in the tourism resource database and optimize the parameters of the recommendation algorithm. 8.The city-level tourist attraction-oriented smart tourism service method of claim 3, wherein, the feature vector of the attraction i at least comprises: User-item matching features for quantifying the fit between user preferences and attraction attributes ; Itinerary constraint features for ensuring that itineraries are feasible in terms of time, budget, and geography ; Context and sequence features for optimizing quality of experience and logical coherence of a journey ; wherein the user-item matching feature by the user portrait vector with the portrait vector of the attraction i The cosine similarity between the two is evaluated, denoted as: ; The travel constraint feature including a moving cost from a previous attraction to the candidate attraction i , a tour time cost , a tour money cost , a geographical distance, a time availability and a budget availability ; The context with sequence features comprises an indicator variable of whether the candidate POI i is in the same region as the previous POI, the category diversity entropy of the itinerary after adding the candidate POI, the normalized popularity of the candidate POI, and a fatigue index representing the current fatigue level of the user. The category diversity entropy cd is calculated by the following formula: ; C is a set of point types, is the proportion of points of type c in the itinerary after adding the candidate point. the fatigue index is calculated by the following equation: ; is a preset intensity tag value of the attraction i, is a maximum value of the preset intensity tag. 9.The city-level tourist attraction-oriented smart tourism service method according to any one of claims 1-8, characterized in that, The method provides services to users through a web application or app developed based on a Vue.js front-end framework. 10.A smart tourism service platform for urban-level tourist attractions, characterized in that, A smart tourism service method for city-level tourist attractions, including: A travel planning module for generating personalized initial travel planning schemes; A booking and payment module for completing one-stop online booking and payment and generating electronic vouchers; An LBS guide module for triggering and pushing multimedia guide content based on user location; A dynamic recommendation module for dynamically adjusting and pushing recommendation information during the tour; A data collection and feedback module for receiving user evaluations and travelogues and collecting data; A data management module for storing, integrating and standardizing tourism resource data and performing algorithm optimization iteration.

Images