A personalized travel itinerary recommendation system and method

Abstract

This invention provides a personalized travel itinerary recommendation system and method. The system includes a user profile construction module, a tourism resource semantic modeling module, an itinerary generation module, an itinerary dynamic optimization module, and an interactive feedback module. The user profile construction module is used to construct a user interest profile. The tourism resource semantic modeling module is used to construct a tourism resource semantic model. The itinerary generation module is used to filter a set of candidate tourism resources that match user preferences based on the matching relationship between the user interest profile and the tourism resource semantic model, and generate a personalized itinerary planning scheme that meets the user's needs. The itinerary dynamic optimization module is used to dynamically adjust and optimize the original itinerary plan during the execution of the travel itinerary. The interactive feedback module is used to update and adjust the user interest profile and subsequent itinerary planning scheme based on user feedback information. This invention can generate and continuously optimize personalized travel itinerary planning schemes based on user interests and real-time travel status, thereby improving the accuracy, flexibility, and overall user experience of travel recommendation results.

Description

Technical Field

[0001] This invention relates to the field of information recommendation system technology, and in particular to a personalized travel itinerary recommendation system and method. Background Technology

[0002] With the rapid development of internet technology and mobile terminals, online travel service platforms have gradually become an important tool for users to obtain travel information and make travel plans. Existing travel recommendation systems usually provide users with travel references by displaying attraction information, user reviews or popular lists, but such methods often lack in-depth analysis of individual user interests and preferences, making it difficult to generate truly personalized travel itinerary plans that meet user needs.

[0003] In actual travel, different users have significant differences in their travel theme preferences, travel modes, and itinerary pace. Existing travel recommendation methods are often based on simple interest tag matching or popular recommendation strategies for itinerary planning, which makes it difficult to comprehensively consider multiple factors such as user interests, tourism resource characteristics, and spatial location relationships, resulting in insufficient personalization of recommendation results.

[0004] In addition, users may be affected by factors such as changes in traffic conditions, adjustments to scenic spot opening hours, and changes in their own itineraries during actual travel, making it difficult for the original travel itinerary plan to remain applicable. Moreover, most existing travel recommendation systems lack the ability to dynamically adjust the itinerary after it is generated, making it difficult to optimize the travel itinerary based on real-time environmental changes.

[0005] Therefore, how to effectively organize tourism resource information based on a thorough analysis of user interest characteristics, and generate more reasonable tourism itinerary planning schemes in conjunction with real-time changes during the trip execution process, has become an urgent problem to be solved in the field of tourism recommendation technology.

[0006] A review of publicly available technical solutions reveals that CN116089730B proposes a method for recommending travel itineraries. This method includes: obtaining user-inputted travel date, number of travel days, departure city, transit cities, arrival city, itinerary recommendation type, and itinerary preference type; determining a target city itinerary based on a pre-built tourism knowledge graph, considering the number of travel days, departure city, transit cities, and arrival city; determining a node recommendation score for each itinerary node based on the pre-built tourism knowledge graph, considering the itinerary preference type, itinerary recommendation type, travel date, and the average sales ratio of itinerary nodes associated with each city node in the target city itinerary; and determining a recommended itinerary route based on the pre-built tourism knowledge graph and the node recommendation score, and then pushing the recommended itinerary route to the user. However, this solution lacks continuous modeling of real-time changes in user interests and individual behavioral characteristics, and it is difficult to adjust the itinerary in real-time during actual travel, thus exhibiting certain deficiencies in personalization and itinerary adaptability. Summary of the Invention

[0007] The purpose of this invention is to address the shortcomings of current methods by proposing a personalized travel itinerary recommendation system and method.

[0008] The present invention adopts the following technical solution: A personalized travel itinerary recommendation system includes a user profile building module, a tourism resource semantic modeling module, an itinerary generation module, an itinerary dynamic optimization module, and an interactive feedback module. The user profile building module collects user travel-related behavioral data and constructs a user interest profile. The tourism resource semantic modeling module collects and stores tourism resource information within a user-specified travel area and constructs a tourism resource semantic model. The itinerary generation module filters a set of candidate tourism resources that match user preferences based on the matching relationship between the user interest profile and the tourism resource semantic model, and generates a personalized itinerary plan that meets the user's needs. The itinerary dynamic optimization module dynamically adjusts and optimizes the original itinerary plan during the travel itinerary execution process. The interactive feedback module updates and adjusts the user interest profile and subsequent itinerary plans based on user feedback.

[0009] The user profile building module includes a behavior data collection unit, a user feature extraction unit, and an interest profile generation unit. The behavior data collection unit collects user travel-related behavior data, including browsing history, search history, historical travel records, stay preference records, and user-initiated preference settings related to each travel theme. The user feature extraction unit cleans, classifies, and extracts features from the collected behavior data, and analyzes the user's behavioral contribution features on each travel theme based on the user behavior type and the travel theme category to which the behavior content belongs. The interest profile generation unit constructs user interest profiles based on the user's behavioral contribution features.

[0010] Furthermore, the tourism resource semantic modeling module includes a resource data acquisition unit, a resource information processing unit, and a semantic tag modeling unit. The resource data acquisition unit is used to collect tourism resource information within a user-specified travel area. This tourism resource information includes scenic spot resources, accommodation resources, catering resources, transportation resources, evaluation resource information, and time-of-use opening information. The resource information processing unit is used to deduplicate, classify, standardize, and structurally organize the collected tourism resource information. The semantic tag modeling unit is used to establish semantic tags for the processed tourism resource information based on the category attributes, theme attributes, geographical attributes, time attributes, and service attributes of tourism resources, thereby forming a tourism resource semantic model that can be used for matching analysis.

[0011] Furthermore, the itinerary generation module includes a resource matching calculation unit, a candidate resource filtering unit, and an itinerary path generation unit. The resource matching calculation unit is used to obtain the interest matching index of each tourism resource based on the matching relationship between the user's interest profile and the tourism resource semantic model. The candidate resource filtering unit is used to sort the tourism resources according to the interest matching index and select tourism resources that meet the preset matching threshold to construct a candidate tourism resource set. The itinerary path generation unit is used to generate a personalized tourism itinerary planning scheme that meets the user's needs based on the candidate tourism resource set and combined with the user's itinerary constraints.

[0012] Furthermore, the itinerary dynamic optimization module includes a planning adjustment judgment unit and a route replanning unit; the planning adjustment judgment unit is used to obtain relevant information that triggers route replanning during the execution of the travel itinerary, and to determine whether the current travel itinerary plan needs to be adjusted based on the obtained information; the route replanning unit is used to dynamically adjust and optimize the original travel itinerary plan when the system detects that the current travel itinerary needs to be replanned.

[0013] Furthermore, the interactive feedback module includes a feedback information collection unit and a profile update unit; the feedback information collection unit is used to obtain user feedback information during the generation or execution of travel itinerary planning; the interest profile update unit is used to update the user interest profile based on the user feedback information.

[0014] Furthermore, the specific calculation method for the interest matching index of each tourism resource is as follows: ; in, For the first The interest matching index between a tourism resource and a user indicates the user's overall level of interest in that tourism resource. For the first A spatial interest modulation function for a tourism resource is used to characterize the impact of the spatial distance between the tourism resource and the user's current location on the degree of matching of the user's interests. This indicates that user interest profiles, obtained through user interest profiling, indicate the user's interest in the first... Interest vectors for tourism-related themes This indicates the first [item] obtained through the tourism resource semantic model. The first tourism resource in the Suitability in similar tourism themes The number of preset tourism theme categories must satisfy the following: ; in, For the first The spatial distance between tourism resources and users The basic spatial radius of influence of user interests is used to characterize the maximum range of interest influence that a user can accept during the current trip planning phase.

[0015] A personalized travel itinerary recommendation method, applied to a personalized travel itinerary recommendation system, the method comprising the following steps: S1: Collect user behavior data related to travel, and perform feature extraction and interest modeling on the collected behavior data to build user interest profiles; S2: Collect and organize tourism resource information within the user-specified travel area, and establish tourism resource semantic tags based on the category attributes, theme attributes, geographical attributes, time attributes, and service attributes of tourism resources, thereby constructing a tourism resource semantic model; S3: Calculate the tourism resource interest matching index based on the matching relationship between the user interest profile and the tourism resource semantic model, and select a set of candidate tourism resources that meet the user's preferences based on the matching index. Based on the set of candidate tourism resources and combined with the user's itinerary constraints, generate a personalized tourism itinerary planning scheme. S4: When the current travel itinerary needs to be replanned during the execution of the travel itinerary, the dynamic matching index of the remaining travel resources is recalculated based on the user's current location, remaining travel time and changes in the travel environment. The set of travel resources that can be replanned is then selected based on the dynamic matching index, thereby dynamically adjusting and optimizing the original travel itinerary plan and generating an updated travel itinerary planning path. S5: Obtain user feedback information during trip planning or trip execution, update the user interest profile based on the user feedback information, and adjust subsequent travel itinerary planning schemes to ensure that the recommendation results continue to match the user's latest interests and preferences.

[0016] The beneficial effects achieved by this invention are: This solution collects user behavior data related to travel, extracts features, and models interests to construct user interest profiles that reflect user preferences, enabling the system to accurately depict user interests across different travel themes. By uniformly organizing tourism resource information within the travel area and establishing semantic tags, the characteristics of each tourism resource across different travel themes can be structurally expressed, providing a unified data foundation for matching analysis between user interests and tourism resources. By calculating a tourism resource interest matching index and introducing a spatial interest modulation mechanism during the index calculation process, the spatial distance between tourism resources and the user's current location has a reasonable impact on the matching results, achieving collaborative matching between interest preferences and spatial accessibility, thus improving the rationality of itinerary planning results. Furthermore, by dynamically adjusting the original itinerary plan based on the user's current location, remaining travel time, and changes in the travel environment during the travel itinerary execution, the itinerary plan can be optimized in real time according to the user's actual travel status, improving the flexibility and executability of the itinerary planning. Attached Figure Description

[0017] The invention will be further understood from the following description taken in conjunction with the accompanying drawings. The components in the drawings are not necessarily drawn to scale, but rather the emphasis is on illustrating the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.

[0018] Figure 1 This is a schematic diagram of the overall modules of the present invention.

[0019] Figure 2 This is a schematic diagram of the personalized travel itinerary recommendation method of the present invention.

[0020] Figure 3 This is a schematic diagram comparing the system of this invention with the traditional system in terms of key indicators.

[0021] Figure 4This diagram illustrates the improvement of the system of the present invention in various key indicators compared to the traditional system. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to its embodiments. It should be understood that the specific embodiments described herein are only for explaining the present invention and are not intended to limit the present invention. Other systems, methods, and / or features of this embodiment will become apparent to those skilled in the art after reviewing the following detailed description. It is intended that all such additional systems, methods, features, and advantages are included within this specification, are included within the scope of the present invention, and are protected by the appended claims. Further features of the disclosed embodiments are described in the following detailed description, and these features will be apparent from the following detailed description.

[0023] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or component referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present patent. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0024] Example 1

[0025] like Figure 1 As shown in the figure, this embodiment provides a personalized travel itinerary recommendation system. The system includes a user profile construction module, a tourism resource semantic modeling module, an itinerary generation module, an itinerary dynamic optimization module, and an interactive feedback module. The user profile construction module is used to collect user travel-related behavioral data and construct a user interest profile. The tourism resource semantic modeling module is used to collect and store tourism resource information within the travel area specified by the user and construct a tourism resource semantic model. The itinerary generation module is used to filter out a set of candidate tourism resources that meet the user's preferences based on the matching relationship between the user interest profile and the tourism resource semantic model, and generate a personalized itinerary planning scheme that meets the user's needs. The itinerary dynamic optimization module is used to dynamically adjust and optimize the original itinerary plan during the execution of the travel itinerary. The interactive feedback module is used to update and adjust the user interest profile and subsequent itinerary planning scheme based on user feedback information. The user profile building module includes a behavior data collection unit, a user feature extraction unit, and an interest profile generation unit. The behavior data collection unit collects user travel-related behavior data, including browsing history, search history, historical travel records, stay preference records, and user-initiated preference settings related to each travel theme. The user feature extraction unit cleans, classifies, and extracts features from the collected behavior data, and analyzes the user's behavioral contribution characteristics on each travel theme based on the user's behavior type and the travel theme category to which the behavior content belongs. The interest profile generation unit constructs user interest profiles based on the user's behavioral contribution characteristics. Furthermore, the user feature extraction unit categorizes user behavior data according to travel theme categories and maps each behavior record in the behavior data to the corresponding travel theme category. It then calculates the contribution value of each behavior record to each travel theme category to characterize the user's behavioral contribution characteristics on each travel theme. The contribution value is calculated as follows: ; in, Indicates the first The first user behavior record The contribution value of travel-related themes, Indicates the first The behavior type weights corresponding to each user behavior record; the behavior type weights are preset according to the user behavior type, such as browsing behavior, search behavior, favorite behavior, historical booking behavior and stay behavior, and different weight values ​​are set for each behavior type according to the degree of fit between each behavior type and the specific travel process. Indicates the first User behavior records and the first The matching degree between travel themes is calculated based on the semantic matching degree between travel resource tags and travel theme categories in user behavior records; Furthermore, the user interest profile constructed by the interest profile generation unit is represented by the following set of user interest vectors: ; in, Represents the set of user interest vectors. Indicates the user's opinion on the first Interest vectors for tourism-related themes The number of preset tourism theme categories includes natural landscapes, historical and cultural sites, culinary experiences, leisure and entertainment, shopping and consumption, and family-friendly experiences, etc. satisfy: ; in, This is the time decay coefficient, used to control the degree of influence of historical behavior on the current interest vector. Its preferred value range is... ; For the first The time difference between the occurrence of the user behavior record and the current time. The number of user behavior records; The tourism resource semantic modeling module includes a resource data acquisition unit, a resource information processing unit, and a semantic tag modeling unit. The resource data acquisition unit collects tourism resource information within a user-specified travel area. This information includes scenic spot resources, accommodation resources, catering resources, transportation resources, evaluation information, and time-of-use information. The resource information processing unit deduplicates, categorizes, standardizes, and structures the collected tourism resource information. The semantic tag modeling unit establishes semantic tags for the processed tourism resource information based on its category, theme, geographic, temporal, and service attributes, forming a tourism resource semantic model that can be used for matching analysis. Furthermore, the semantic tag modeling unit calculates the suitability of each tourism resource across different travel theme dimensions based on the semantic matching relationship between tourism resource semantic tags and preset travel theme categories, thereby constructing a set of tourism resource theme suitability vectors to be stored in a tourism resource semantic model that can be used for matching analysis; the tourism resource theme suitability vector is represented as: ; in, Indicates the first The theme fit vector corresponding to each tourism resource; Indicates the first The first tourism resource in the The fit degree on tourism themes is calculated based on the semantic matching degree between the tourism resource tag set and the travel theme tag set, so as to characterize the feature strength of the tourism resource in each travel theme dimension. The itinerary generation module includes a resource matching calculation unit, a candidate resource filtering unit, and an itinerary path generation unit. The resource matching calculation unit is used to obtain the interest matching index of each tourism resource based on the matching relationship between the user's interest profile and the tourism resource semantic model. The candidate resource filtering unit is used to sort the tourism resources according to the interest matching index and select tourism resources that meet the preset matching threshold to construct a candidate tourism resource set. The itinerary path generation unit is used to generate a personalized tourism itinerary planning scheme that meets the user's needs based on the candidate tourism resource set and combined with the user's itinerary constraints. Furthermore, the specific calculation method for the interest matching index of each tourism resource is as follows: ; in, For the first The interest matching index between a tourism resource and a user indicates the user's overall level of interest in that tourism resource. For the first A spatial interest modulation function for a tourism resource is used to characterize the impact of the spatial distance between the tourism resource and the user's current location on the degree of matching the user's interests; satisfying: ; in, For the first The spatial distance between tourism resources and users The basic spatial radius of influence of user interests is used to characterize the maximum range of interest influence that the user can accept during the current trip planning stage. It is preset based on the spatial distance between the center of the travel area and the boundary of the area. Furthermore, the candidate resource screening unit sorts each tourism resource in descending order according to the interest matching index of each tourism resource, and selects tourism resources that meet the conditions and includes them in the candidate tourism resource set according to the preset matching threshold and quantity limit. Furthermore, the itinerary route generation unit first obtains the itinerary constraints input by the user, including information such as the number of travel days, start and end times, budget range, and mode of transportation, and determines the starting node of the itinerary based on the user's current location or the starting point of the itinerary. Then, based on the set of candidate tourist resources, it constructs a tourist resource path network according to the spatial relationships and transportation distances between the tourist resources, and generates several candidate itinerary paths using existing path planning methods. The candidate itinerary path generation method can employ a path combination approach based on the Traveling Salesman Problem, or a heuristic path search method to combine candidate tourist resources to obtain multiple sets of candidate itinerary paths. After generating the candidate itinerary paths, the itinerary route generation unit combines the various tourist resources... The system assesses the time feasibility of each candidate travel route based on the estimated tour duration, scenic area opening hours, and the user's available travel time. It calculates the total tour time and total travel time for each candidate route and compares this with the user's travel time constraints, eliminating candidate routes that do not meet the time constraints. Based on this, the travel route generation unit comprehensively evaluates the remaining candidate routes. This comprehensive evaluation can be based on a multi-factor index approach, considering factors such as total route distance, total tour time, interest matching index, and the degree of matching with scenic area opening hours. The priority of candidate routes is determined through weighted indexes or ranking methods, and the travel route with the highest score or that meets the preset optimization objective is selected as the final personalized travel itinerary planning scheme output. This solution collects, extracts features from, and models user behavior data related to travel, and uses a time decay mechanism to weight user historical behavior. This allows for a more accurate portrayal of users' current travel interests and preferences, enabling the generated user interest profiles to dynamically reflect changing trends and improve the accuracy of user interest identification. Furthermore, it calculates interest matching indices for various tourism resources to match user interests with the semantic features of those resources. A spatial interest modulation function reflects the impact of spatial distance between tourism resources and the user's current location on the degree of interest matching. This achieves tourism resource interest matching under the combined influence of time and spatial factors, improving the rationality and feasibility of generated travel itineraries. The interactive feedback module includes a feedback information collection unit and a profile update unit. The feedback information collection unit is used to acquire user feedback information during the generation or execution of travel itinerary planning. The user feedback information includes user click behavior, collection behavior, evaluation behavior, itinerary adjustment operation, and user-initiated preference modification information. The interest profile update unit is used to update the user interest profile based on the user feedback information. By combining the user feedback information with the user feature extraction unit and the interest profile generation unit, the user interest profile is updated and constructed, so that the user interest profile can continuously reflect the user's current interest and preference changes. like Figure 2 As shown, a personalized travel itinerary recommendation method is applied to a personalized travel itinerary recommendation system. The method includes the following steps: S1: Collect user behavior data related to travel, and perform feature extraction and interest modeling on the collected behavior data to build user interest profiles; S2: Collect and organize tourism resource information within the user-specified travel area, and establish tourism resource semantic tags based on the category attributes, theme attributes, geographical attributes, time attributes, and service attributes of tourism resources, thereby constructing a tourism resource semantic model; S3: Calculate the tourism resource interest matching index based on the matching relationship between the user interest profile and the tourism resource semantic model, and select a set of candidate tourism resources that meet the user's preferences based on the matching index. Based on the set of candidate tourism resources and combined with the user's itinerary constraints, generate a personalized tourism itinerary planning scheme. S4: When the current travel itinerary needs to be replanned during the execution of the travel itinerary, the dynamic matching index of the remaining travel resources is recalculated based on the user's current location, remaining travel time and changes in the travel environment. The set of travel resources that can be replanned is then selected based on the dynamic matching index, thereby dynamically adjusting and optimizing the original travel itinerary plan and generating an updated travel itinerary planning path. S5: Obtain user feedback information during trip planning or trip execution, update the user interest profile based on the user feedback information, and adjust subsequent travel itinerary planning schemes to ensure that the recommendation results continue to match the user's latest interests and preferences.

[0026] Example 2

[0027] This embodiment should be understood to include at least all the features of any of the foregoing embodiments, and to further improve upon them; This embodiment provides a personalized travel itinerary recommendation system. The system includes a user profile construction module, a tourism resource semantic modeling module, an itinerary generation module, an itinerary dynamic optimization module, and an interactive feedback module. The user profile construction module collects user travel-related behavioral data and constructs a user interest profile. The tourism resource semantic modeling module collects and stores tourism resource information within a user-specified travel area and constructs a tourism resource semantic model. The itinerary generation module filters a set of candidate tourism resources that match user preferences based on the matching relationship between the user interest profile and the tourism resource semantic model, and generates a personalized itinerary planning scheme that meets the user's needs. The itinerary dynamic optimization module dynamically adjusts and optimizes the original itinerary plan during the execution of the travel itinerary. The interactive feedback module updates and adjusts the user interest profile and subsequent itinerary planning scheme based on user feedback information. Furthermore, the itinerary dynamic optimization module includes a planning adjustment judgment unit and a route replanning unit; the planning adjustment judgment unit is used to obtain relevant information that triggers route replanning during the execution of the travel itinerary, and to determine whether the current travel itinerary plan needs to be adjusted based on the obtained information; the route replanning unit is used to dynamically adjust and optimize the original travel itinerary plan when the system detects that the current travel itinerary needs to be replanned. Furthermore, the relevant information for triggering route replanning includes tourism environment change information and user-initiated trigger information, wherein the tourism environment change information includes changes in the opening status of scenic spots, changes in the level of crowding in scenic spots, changes in traffic conditions, and weather changes, etc.; when the planning adjustment judgment unit detects that the tourism environment change information does not meet the user's preset travel conditions or user-initiated trigger information, it triggers route replanning. Furthermore, during the dynamic adjustment and optimization of the original travel itinerary plan, the route replanning unit uses the user's current location as the new starting node of the itinerary and re-evaluates the remaining travel resources, calculating the dynamic matching index of the remaining travel resources: ; in, For the first The remaining tourism resources are in the current redevelopment period. The dynamic matching index at any given time represents the overall attractiveness of the tourism resource to users at the current stage of their trip. This is a dynamic spatial modulation function used to characterize the influence of spatial distance and accessibility of tourism resources on the matching index; it satisfies: ; in, The average speed corresponding to the user's current mode of transportation. For users' remaining disposable travel time, For the first The spatial distance between remaining tourism resources and users For the first The tourism category sensitivity coefficient corresponding to each remaining tourism resource is used to characterize the sensitivity of different categories of tourism resources to spatial distance, and the preferred value range is [value range missing]. The value is set based on the typical access distance range of different types of tourism resources. Tourism resource categories with higher distance sensitivity correspond to smaller values, while those with lower distance sensitivity correspond to larger values. For example, for the following tourism resource categories: Catering resources and convenience service resources: ; Leisure and entertainment resources: ; Core scenic areas or landmark attractions: ; Furthermore, after obtaining the dynamic matching index of each remaining travel resource, the route replanning unit sorts the remaining travel resources according to a preset screening rule, selects the travel resources with a dynamic matching index higher than a preset threshold as the set of replannable travel resources, and inputs the current set of replannable travel resources into the itinerary route generation unit. Taking the user's current location as the new itinerary starting node, and combining the current itinerary constraints, a new travel itinerary planning scheme is output to complete the replanning of the itinerary route. This solution addresses the need for itinerary replanning during travel by acquiring the user's current location, remaining travel time, and information on changes in the travel environment. It then recalculates the dynamic matching index for remaining travel resources, filters out replannable resources based on this index, and uses the user's current location as a new starting point to rearrange routes and travel times for the remaining resources. This generates an updated itinerary plan, allowing for dynamic adjustments based on the user's current travel status. Furthermore, by introducing remaining time constraints and a sensitivity coefficient for different types of travel resources during route replanning, the spatial impact of different types of travel resources is appropriately adjusted. This results in updated itineraries that better reflect actual travel behavior, improving the flexibility, rationality, and overall user experience of the itinerary planning.

[0028] Example 3

[0029] This embodiment should be understood to include at least all the features of any of the foregoing embodiments, and to further improve upon them; This embodiment provides a personalized travel itinerary recommendation system. The system includes a user profile construction module, a tourism resource semantic modeling module, an itinerary generation module, an itinerary dynamic optimization module, and an interactive feedback module. During the actual operation of the system, each module is executed sequentially according to a preset data processing flow, thereby completing operations such as user interest analysis, tourism resource matching, and itinerary planning generation. After the system starts, the user profile building module first collects and processes user behavior data related to tourism. Specifically, the behavior data collection unit can obtain user browsing history, search history, favorites history, historical booking history, and user dwell time through mobile terminal applications or online travel platforms. After obtaining the above behavior data, the user feature extraction unit cleans the behavior data, such as removing duplicate data, filtering abnormal data, and standardizing the behavior data format. Then, it classifies user behavior according to the tourism resource tags and travel theme categories involved in the behavior records, thereby extracting user behavior features on different travel themes. Subsequently, the interest profile generation unit constructs user interest profiles based on the extracted behavior features, enabling the system to obtain the user's interest tendencies on different travel theme dimensions. After constructing the user interest profile, the system uses a tourism resource semantic modeling module to uniformly collect and organize tourism resource information within the user's designated travel area. Specifically, the resource data collection unit can obtain tourism resource information through tourism platform databases, scenic spot information interfaces, or public tourism information platforms. This tourism resource information includes scenic spot resources, accommodation resources, catering resources, transportation resources, and tourism evaluation information. After acquiring the tourism resource data, the resource information organization unit classifies, deduplicates, and standardizes the collected data, and stores it according to a unified data structure. Subsequently, the semantic tag modeling unit establishes semantic tags for each tourism resource based on its category attributes, theme attributes, geographical location attributes, and opening time attributes, thereby forming a tourism resource semantic model that can be used for matching analysis. After the user interest profile and tourism resource semantic model are established, the system enters the itinerary generation stage. Specifically, the resource matching calculation unit calculates the interest matching index of each tourism resource based on the matching relationship between the user interest profile and the tourism resource semantic model, and sorts the tourism resources according to the interest matching index. Subsequently, the candidate resource screening unit selects tourism resources that meet the user's preferences from the sorting results according to a preset matching threshold and constructs a candidate tourism resource set. After obtaining the candidate tourism resource set, the itinerary path generation unit combines the candidate tourism resources according to the itinerary constraints input by the user, thereby generating a personalized tourism itinerary planning scheme that meets the user's needs. Specifically, when generating a travel itinerary, the itinerary generation unit first obtains the itinerary information input by the user, such as the number of travel days, daily sightseeing time range, budget constraints, and mode of transportation, and uses the user's current location or the starting location of the itinerary as the starting point of the path; then, it constructs a travel resource path network based on the spatial relationship between candidate travel resources and the transportation distance, and generates multiple candidate itinerary paths through path combination; after generating candidate paths, the system further combines the opening hours of scenic spots, the estimated sightseeing time, and the user's remaining available time to make a feasibility judgment on each candidate path, and eliminates paths that do not meet the time constraints, thereby obtaining the final personalized travel itinerary planning scheme; During the user's trip according to the generated itinerary plan, the system continuously acquires information related to the tourism environment through the itinerary dynamic optimization module, such as changes in the opening status of scenic spots, changes in traffic conditions, weather changes, and information on user-initiated itinerary adjustments. When the system detects that the above information may affect the original itinerary plan, the planning adjustment judgment unit evaluates the current itinerary plan. If it determines that the itinerary needs to be replanned, the route replanning process is triggered. During the route replanning process, the system uses the user's current location as the new itinerary starting point and recalculates the matching index of the remaining tourism resources to generate a new tourism itinerary planning path. In addition, during the user's use of the system, the interactive feedback module continuously collects user feedback information, such as user click behavior, collection behavior, evaluation behavior of attractions, and information such as user's active modification of itinerary arrangements; the interest profile update unit updates the user's interest profile according to the above feedback information, so that the system can continuously learn the changes in user's interests, thereby generating a travel itinerary planning scheme that is more in line with user preferences in the subsequent recommendation process. like Figure 3 , Figure 4 As shown, to verify the actual effect of the personalized travel itinerary recommendation system of the present invention, 500 users with real travel planning needs were selected for testing, and the technical solution of this invention was compared and analyzed with traditional travel recommendation systems. The traditional system mainly generates travel itinerary plans based on popular attraction recommendations or simple interest tag matching, while the technical solution of this invention generates personalized itinerary plans through user interest profiling, semantic modeling of tourism resources, and dynamic itinerary optimization mechanisms. Regarding recommendation accuracy, by analyzing user clicks and adoption rates for recommended attractions, this technical solution achieved a 92.5% accuracy rate, compared to 74.3% for traditional recommendation systems, representing an improvement of approximately 24.5%. In terms of itinerary completion rate, by analyzing the percentage of users actually completing the recommended itinerary, this technical solution achieved a 90.8% completion rate, compared to 71.6% for traditional systems, representing an improvement of approximately 26.8%. Regarding user satisfaction, user ratings showed an average satisfaction score of 4.6 out of 5 for this technical solution, compared to 3.8 for traditional recommendation systems, representing an improvement of approximately 21%. Finally, regarding dynamic itinerary adjustment capabilities, by simulating environmental changes such as scenic area congestion and traffic variations, this technical solution successfully completed dynamic itinerary adjustments in 88.2% of cases, compared to only 52.4% for traditional systems, representing an improvement of approximately 68%. Experimental results show that this technical solution is significantly better than traditional travel recommendation systems in key indicators such as recommendation accuracy, itinerary completion rate, user satisfaction, and dynamic itinerary adjustment capability, thus providing users with more accurate, flexible, and personalized travel itinerary planning solutions that match their interests and preferences.

[0030] The content disclosed above is only a preferred and feasible embodiment of the present invention, and is not intended to limit the scope of protection of the present invention. Therefore, all equivalent technical changes made based on the content of the present invention specification and drawings are included within the scope of protection of the present invention. Furthermore, the elements therein can be updated as technology develops.

Claims

1. A personalized travel itinerary recommendation system, characterized in that, The system includes a user profile building module, a tourism resource semantic modeling module, an itinerary generation module, an itinerary dynamic optimization module, and an interactive feedback module. The user profile building module collects user travel-related behavioral data and constructs user interest profiles. The tourism resource semantic modeling module collects and stores tourism resource information within a user-specified travel area and constructs a tourism resource semantic model. The itinerary generation module filters candidate tourism resources that match user preferences based on the matching relationship between the user interest profile and the tourism resource semantic model, and generates a personalized itinerary planning scheme that meets the user's needs. The itinerary dynamic optimization module dynamically adjusts and optimizes the original itinerary plan during the execution of the travel itinerary. The interactive feedback module updates and adjusts the user interest profile and subsequent itinerary planning scheme based on user feedback. The user profile building module includes a behavior data collection unit, a user feature extraction unit, and an interest profile generation unit. The behavior data collection unit collects user travel-related behavior data, including browsing history, search history, historical travel records, stay preference records, and user-initiated preference settings related to each travel theme. The user feature extraction unit cleans, classifies, and extracts features from the collected behavior data, and analyzes the user's behavioral contribution features on each travel theme based on the user behavior type and the travel theme category to which the behavior content belongs. The interest profile generation unit constructs user interest profiles based on the user's behavioral contribution features.

2. The personalized travel itinerary recommendation system according to claim 1, characterized in that, The tourism resource semantic modeling module includes a resource data acquisition unit, a resource information processing unit, and a semantic tag modeling unit. The resource data acquisition unit is used to collect tourism resource information within a user-specified travel area. This tourism resource information includes scenic spot resources, accommodation resources, catering resources, transportation resources, evaluation resource information, and time-of-use opening information. The resource information processing unit is used to deduplicate, classify, standardize, and structure the collected tourism resource information. The semantic tag modeling unit is used to establish semantic tags for the processed tourism resource information based on the category attributes, theme attributes, geographical attributes, time attributes, and service attributes of tourism resources, thereby forming a tourism resource semantic model that can be used for matching analysis.

3. The personalized travel itinerary recommendation system according to claim 1, characterized in that, The itinerary generation module includes a resource matching calculation unit, a candidate resource filtering unit, and an itinerary path generation unit. The resource matching calculation unit is used to obtain the interest matching index of each tourism resource based on the matching relationship between the user's interest profile and the tourism resource semantic model. The candidate resource filtering unit is used to sort the tourism resources according to the interest matching index and select tourism resources that meet the preset matching threshold to construct a candidate tourism resource set. The itinerary path generation unit is used to generate a personalized tourism itinerary planning scheme that meets the user's needs based on the candidate tourism resource set and combined with the user's itinerary constraints.

4. The personalized travel itinerary recommendation system according to claim 1, characterized in that, The itinerary dynamic optimization module includes a planning adjustment determination unit and a route replanning unit; the planning adjustment determination unit is used to obtain relevant information that triggers route replanning during the execution of the travel itinerary, and to determine whether the current travel itinerary plan needs to be adjusted based on the obtained information. The route replanning unit is used to dynamically adjust and optimize the original travel itinerary plan when the system detects that the current travel itinerary needs to be replanned.

5. A personalized travel itinerary recommendation system according to claim 1, characterized in that, The interactive feedback module includes a feedback information collection unit and a profile update unit; the feedback information collection unit is used to obtain user feedback information during the generation or execution of travel itinerary planning; the interest profile update unit is used to update the user interest profile based on the user feedback information.

6. The personalized travel itinerary recommendation system according to claim 3, characterized in that, The specific calculation method for the interest matching index of each tourism resource is as follows: Sj=∅j∙i=1nμi∙rji; Where Sj is the interest matching index between the j-th tourism resource and the user, representing the user's overall tourism interest in the tourism resource; ∅j is the spatial interest modulation function of the j-th tourism resource, used to characterize the influence of the spatial distance between the tourism resource and the user's current location on the user's interest matching degree; ui represents the user's interest vector for the i-th type of tourism theme obtained through user interest profile; rji represents the fit degree of the j-th tourism resource on the i-th type of tourism theme obtained through tourism resource semantic model; n is the preset number of tourism theme categories, satisfying: ∅j=exp(−djR); Where dj is the spatial distance between the j-th tourism resource and the user, and R is the basic spatial radius of influence of the user's interest, which is used to characterize the maximum range of interest influence that the user can accept in the current trip planning stage.

7. A personalized travel itinerary recommendation method, applied to the personalized travel itinerary recommendation system described in claim 1, characterized in that, The method includes the following steps: S1: Collect user behavior data related to travel, and perform feature extraction and interest modeling on the collected behavior data to build user interest profiles; S2: Collect and organize tourism resource information within the user-specified travel area, and establish tourism resource semantic tags based on the category attributes, theme attributes, geographical attributes, time attributes, and service attributes of tourism resources, thereby constructing a tourism resource semantic model; S3: Calculate the tourism resource interest matching index based on the matching relationship between the user interest profile and the tourism resource semantic model, and select a set of candidate tourism resources that meet the user's preferences based on the matching index. Based on the set of candidate tourism resources and combined with the user's itinerary constraints, generate a personalized tourism itinerary planning scheme. S4: When the current travel itinerary needs to be replanned during the execution of the travel itinerary, the dynamic matching index of the remaining travel resources is recalculated based on the user's current location, remaining travel time and changes in the travel environment. The set of travel resources that can be replanned is then selected based on the dynamic matching index, thereby dynamically adjusting and optimizing the original travel itinerary plan and generating an updated travel itinerary planning path. S5: Obtain user feedback information during trip planning or trip execution, update the user interest profile based on the user feedback information, and adjust subsequent travel itinerary planning schemes to ensure that the recommendation results continue to match the user's latest interests and preferences.

Images