Travel itinerary recommendation method and system

Abstract

The invention relates to the technical field of information recommendation, in particular to a travel itinerary recommendation method and system. The method comprises the following steps: constructing a user temporary portrait and a group temporary portrait; screening a scenic spot set based on the group portrait, constructing a scenic spot association graph, and generating an initial recommendation path; obtaining current behavior data of multiple users, constructing current portraits of the users, comparing the current portraits of the users with temporary portraits of the users, outputting deviation data, triggering a path updating request, and generating a final recommended path; the system comprises a temporary portrait construction module, an initial recommendation path generation module and a final recommendation path generation module. The initial recommendation path is generated through the temporary user portraits and the temporary group portraits, the current user portraits are constructed according to the current behavior data of the multiple users, the current user portraits and the temporary user portraits are compared to generate the final recommendation path, and path recommendation through real-time behaviors of the multiple users is achieved. And the group journey lack collaboration and the overall tourism experience are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of information recommendation, and particularly relates to a travel itinerary recommendation method and system. BACKGROUND

[0002] The travel recommendation system is an intelligent service system for providing personalized travel itinerary suggestions for users based on user demand, behavior data and environmental information. When traveling, users usually have accompanying personnel. The existing travel recommendation system does not utilize real-time behavior data of accompanying users, resulting in lack of coordination of team itinerary and thus reducing overall travel experience.

[0003] In view of the above, it is necessary to provide a travel itinerary recommendation method and system for improving the lack of coordination of team itinerary and overall travel experience by using real-time behavior of multiple users. SUMMARY

[0004] The present application aims to provide a travel itinerary recommendation method and system, and aims to solve the technical problem that the real-time behavior data of accompanying users is not utilized in the prior art, resulting in lack of coordination of team itinerary and thus reducing overall travel experience.

[0005] To achieve the above-mentioned purpose, a travel itinerary recommendation method adopted by the present application comprises the following steps:

[0006] Respectively acquire multi-user behavior data, social relationship data and environmental data, and construct user temporary portrait and group temporary portrait;

[0007] Filter the scenic spot set based on the group portrait, construct the scenic spot association graph, and generate the initial recommended path;

[0008] Acquire the current behavior data of multiple users, construct the user current portrait, compare the user current portrait and the user temporary portrait, output the deviation data, trigger the path update request, and generate the final recommended path.

[0009] In the step of respectively acquiring multi-user behavior data, social relationship data and environmental data, and constructing user temporary portrait and group temporary portrait:

[0010] Respectively acquire multi-user behavior data, social relationship data and environmental data, and pre-process the data to output the processed data;

[0011] Extract the user interest label according to the processed data, and construct the user temporary portrait;

[0012] According to the processed data, identify the user group and extract the user group features, construct the group temporary portrait, and generate the group-level interest label.

[0013] In the step of extracting user interest tags according to the processed data and constructing a user temporary profile:

[0014] The behavior features are extracted as the user interest tags, and the social features and environmental features are fused to construct the user temporary profile.

[0015] In the step of identifying a user group according to the processed data, extracting user group features, constructing a group temporary profile, and generating group-level interest tags:

[0016] Based on the team ID, the same user is identified, the individual profile in the group is fused, the group temporary profile is constructed, and the group-level interest tags are generated.

[0017] In the step of screening a scenic spot set based on the group profile, constructing a scenic spot association graph, and generating an initial recommended path:

[0018] A scenic spot association graph is constructed through the scenic spot set;

[0019] A path is generated according to the scenic spot association graph, and initial recommended path data is output.

[0020] Before the step of constructing a scenic spot association graph through the scenic spot set:

[0021] A screening threshold is set, and based on the group-level interest tags, a scenic spot set with a matching degree higher than the screening threshold is screened from the scenic spot database.

[0022] In the step of constructing a scenic spot association graph through the scenic spot set:

[0023] The scenic spots are taken as nodes, and the edge weight is the traffic time between the scenic spots, to obtain the scenic spot association graph.

[0024] In the step of obtaining multi-user current behavior data, constructing a user current profile, comparing the user current profile and the user temporary profile, outputting deviation data, triggering a path update request, and generating a final recommended path:

[0025] The user current behavior data is collected in real time, the user current profile is generated, the user current profile and the user temporary profile are compared, the deviation degree is calculated, and the deviation data is output;

[0026] The deviation condition is judged, the path is locally and globally optimized, and the final recommended path is generated.

[0027] In the step of judging the deviation condition, locally and globally optimizing the path, and generating the final recommended path:

[0028] If a single scenic spot deviates, the path is locally optimized;

[0029] If two or more scenic spots deviate, the path is globally optimized.

[0030] The application also provides a travel itinerary recommendation system, comprising a temporary portrait construction module, an initial recommended path generation module, and a final recommended path generation module.

[0031] The temporary portrait construction module is configured to obtain multi-user behavior data, social relationship data, and environmental data respectively, and construct user temporary portraits and group temporary portraits.

[0032] The initial recommended path generation module is configured to filter scenic spot sets based on group portraits, construct a scenic spot association graph, and generate an initial recommended path.

[0033] The final recommended path generation module is configured to obtain multi-user current behavior data, construct user current portraits, compare user current portraits and user temporary portraits, output deviation data, trigger a path update request, and generate a final recommended path.

[0034] The travel itinerary recommendation method and system of the application adopt the temporary portrait construction module, the initial recommended path generation module, and the final recommended path generation module to perform the following steps: obtaining multi-user behavior data, social relationship data, and environmental data respectively, and constructing user temporary portraits and group temporary portraits; filtering scenic spot sets based on group portraits, constructing a scenic spot association graph, and generating an initial recommended path; obtaining multi-user current behavior data, constructing user current portraits, comparing user current portraits and user temporary portraits, outputting deviation data, triggering a path update request, and generating a final recommended path; generating an initial recommended path through user temporary portraits and group temporary portraits, constructing user current portraits according to multi-user current behavior data, comparing user current portraits and user temporary portraits to generate a final recommended path, and realizing path recommendation through real-time behavior of multi-users, improving the lack of collaboration of team itinerary and overall tourism experience. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0036] Figure 1 is a step flow chart of the travel itinerary recommendation method of the application.

[0037] Figure 2 is a step flow chart of S100 of the application.

[0038] Figure 3 This is a flowchart of steps S200 of the present invention.

[0039] Figure 4 This is a flowchart of steps S300 of the present invention.

[0040] Figure 5 This is a schematic diagram of the tourism itinerary recommendation system of the present invention.

[0041] Figure 6 This is a schematic diagram of the electronic device of the present invention.

[0042] 401 - Temporary profile building module, 402 - Initial recommended path generation module, 403 - Final recommended path generation module. Detailed Implementation

[0043] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0044] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0045] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0046] Please see Figures 1-4 This invention provides a method for recommending travel itineraries, comprising the following steps:

[0047] S100: Acquire multi-user behavior data, social relationship data, and environmental data respectively, and construct temporary user profiles and group profiles.

[0048] In this embodiment, multi-user behavior data, social relationship data, and environmental data are acquired respectively, and temporary user profiles and group profiles are constructed. The specific process is as follows:

[0049] S101: Acquire multi-user behavior data, social relationship data, and environmental data respectively, preprocess the data, and output the processed data;

[0050] S102: Extract user interest tags from the processed data, extract behavioral features as user interest tags, and integrate social and environmental features to construct a temporary user profile;

[0051] S103: Identify peer users based on team ID, merge individual profiles within the group, construct a temporary group profile, and generate group-level interest tags.

[0052] In the above process, the data includes user behavior data, social relationship data, and environmental data; among which, user behavior data includes:

[0053] Historical behavior: By querying the database, we can identify users' searches (e.g., "family-friendly hotels"), bookings (attraction tickets), and duration of stay (e.g., browsing an attraction's details page for 5 minutes) over the past 30 days.

[0054] Real-time behavior: The app uses tracking points to capture current actions (such as clicking the "trendy restaurant" tag or swiping to skip the "museum" recommendation).

[0055] Social relationship data:

[0056] Obtain peer relationships (family group tags) and friend preferences (friend A frequently visits the "zoo") from user-authorized social platforms.

[0057] Environmental data:

[0058] Real-time weather (moderate rain in a certain area), tourist attraction crowds (crowding at a certain attraction is 80%), and traffic conditions (congestion on a certain road).

[0059] Preprocessing operations, cleaning: removing duplicate records (such as the same user searching for the same attraction multiple times), and correcting erroneous data (such as correcting "average consumption of 5000 yuan" to "500 yuan").

[0060] Standardization: Unify time format (UTC to local time), normalize values ​​(congestion level mapped to the 0-1 range).

[0061] Noise reduction: Filter low-frequency behaviors (such as accidental clicks on irrelevant attractions) or outliers (dirty data with a dwell time of more than 1 hour).

[0062] Output: A structured data table containing fields such as user ID, behavior type, timestamp, and environment parameters.

[0063] Interest tag extraction and behavioral characteristics: statistical analysis of high-frequency operations (e.g., "searched for 'family hotel' 3 times" → tag "family needs").

[0064] Social features: If a user travels with a "food lover" friend, the weight of the "food exploration" tag will be increased.

[0065] Environmental characteristics: On rainy days, the priority of the "Outdoor Hiking" tag is automatically reduced, and the "Indoor Museum" tag is added.

[0066] User profiles are stored in key-value pairs (e.g., user ID:{tag:weight}) and are valid for 24 hours (dynamically updated).

[0067] Group identification clusters users by team ID (such as family group ID) or behavioral similarity (such as 5 families who simultaneously booked "zoo + children's playground").

[0068] Group profile fusion, common tags: statistical high-frequency tags within the group (e.g., 70% of users have "parent-child needs" → group tag ["parent-child dominant"]).

[0069] Dynamic weighting: Adjust tag priority based on group size (e.g., add "suitable for large groups" tag for a group of 10).

[0070] Group-level tag output is transformed into filtering conditions (such as prioritizing "family-themed hotels" rather than "couples' homestays").

[0071] S200: Based on the group profile, filter the set of attractions, construct the attraction relationship graph, and generate the initial recommended path.

[0072] In this implementation, a set of scenic spots is selected based on the group profile, a scenic spot association graph is constructed, and an initial recommended route is generated. The specific process is as follows:

[0073] S201: Set a filtering threshold and, based on group-level interest tags, filter a set of attractions from the attraction database whose matching degree is higher than the filtering threshold;

[0074] S202: Using attractions as nodes and edge weights equal to the travel time between attractions, obtain an attraction association graph;

[0075] S203: Generate a route based on the attraction association map and output the initial recommended route data.

[0076] In the above process, the screening threshold is set, and the matching threshold (e.g., ≥0.7) is set according to the weight of the group label (e.g., "parent-child needs" weight 0.8).

[0077] Attraction matching: Filter attractions that meet the criteria from the database (e.g., "zoo" matches ["family-oriented"], matching score 0.85).

[0078] Output: A set of attractions {attraction ID:matching degree}, such as {zoo:0.85, children's park:0.8}.

[0079] Node and edge definitions:

[0080] Nodes: Attractions within a collection of attractions (such as zoos and children's parks).

[0081] Edge weight: Travel time between attractions (e.g., according to the map, it takes 15 minutes to drive from "Zoo → Children's Park").

[0082] Output the graph structure to generate a weighted undirected graph G = (V, E), such as V = {Zoo, Children's Park}, E = {(Zoo, Children's Park): 15}.

[0083] Path algorithm selection:

[0084] Use Dijkstra's algorithm to calculate the shortest path (e.g., "Zoo → Children's Park" total time 15 minutes).

[0085] Alternatively, a genetic algorithm can be used to optimize the order of multiple attractions (e.g., “zoo → children’s park → a certain street”, total time 3 hours).

[0086] Output results: Initial recommended route data: [Zoo, Children's Park, A Street], total time: 3 hours.

[0087] S300: Acquire current behavior data of multiple users, construct current user profiles, compare current user profiles with temporary user profiles, output deviation data, trigger path update requests, and generate final recommended paths.

[0088] In this implementation, current behavior data of multiple users is acquired, a current user profile is constructed, the current user profile is compared with a temporary user profile, deviation data is output, a path update request is triggered, and a final recommended path is generated. The specific process is as follows:

[0089] S301: Real-time collection of user's current behavior data, generation of current user profile, comparison of current user profile and temporary user profile, calculation of deviation, and output of deviation data;

[0090] S302: Determine deviations, perform local and global optimizations on the path, and generate the final recommended path.

[0091] Furthermore, in the steps of judging deviations, performing local and global optimizations on the path, and generating the final recommended path:

[0092] If a single attraction deviates from the path, the route will be locally optimized.

[0093] If two or more attractions deviate from each other, the path will be optimized globally.

[0094] During the above process, real-time data is collected to capture the user's current actions (such as user A clicking the "Leave Early" button after staying at the zoo for 30 minutes).

[0095] The current profile is generated, and the tag weights are updated (e.g., the weight of "parent-child needs" decreases from 0.8 to 0.5, and the weight of "avoid crowds" increases to 0.7).

[0096] Deviation is calculated by comparing the label differences between the current image and the temporary image (such as Euclidean distance or cosine similarity).

[0097] Output: Deviation data {User ID:{Deviation label:Change value}}, such as {User A:{"Parent-child needs":-0.3}}.

[0098] Deviation assessment includes deviation from a single attraction and deviation from multiple attractions:

[0099] Individual attraction deviates from its designated location (e.g., user A leaves the zoo early):

[0100] Local optimization: Replace similar attractions in subsequent paths (e.g., replace "Children's Park" with "Indoor Playground").

[0101] Several points of interest were deviated from the designated areas (e.g., two people in the group left the zoo early and one person said they were "hungry"):

[0102] Global optimization: Replan all paths (e.g., insert "food street" and adjust the order).

[0103] Optimization algorithm application

[0104] Local optimization: Use a greedy algorithm to replace a single node (e.g., select the nearest "indoor playground").

[0105] Global optimization: Use simulated annealing algorithm to regenerate the path (e.g., "Zoo → Street → Indoor Amusement Park" takes 2.5 hours in total).

[0106] The final recommended route is: [Zoo (shortened stay), a certain street, indoor playground], with a total time of 2.5 hours.

[0107] In this invention, firstly, multi-user behavior data, social relationship data, and environmental data are acquired, and temporary user profiles and group profiles are constructed. Then, based on the group profiles, a set of attractions is selected, an attraction association graph is constructed, and an initial recommended path is generated. Finally, multi-user current behavior data is acquired, a current user profile is constructed, the current user profile is compared with the temporary user profile, deviation data is output, a path update request is triggered, and a final recommended path is generated. By generating an initial recommended path through the temporary user profiles and group profiles, constructing a current user profile based on the multi-user current behavior data, and comparing the current user profile with the temporary user profile to generate a final recommended path, this invention achieves path recommendation based on the real-time behavior of multiple users, improving the lack of coordination in group itineraries and the overall travel experience.

[0108] Corresponding to the aforementioned embodiments of the travel itinerary recommendation method, this application also provides embodiments of a travel itinerary recommendation system.

[0109] Figure 5 This is a block diagram illustrating a travel itinerary recommendation system according to an exemplary embodiment. (Refer to...) Figure 5 The system may include: a temporary profile construction module 401, an initial recommendation path generation module 402, and a final recommendation path generation module 403; wherein:

[0110] The temporary profile building module 401 is used to acquire multi-user behavior data, social relationship data and environmental data respectively, and to build temporary user profiles and group temporary profiles.

[0111] The initial recommended path generation module 402 is used to filter the set of attractions based on the group profile, construct the attraction association graph, and generate the initial recommended path.

[0112] The final recommended path generation module 403 is used to acquire current behavior data of multiple users, construct current user profiles, compare current user profiles with temporary user profiles, output deviation data, trigger path update requests, and generate final recommended paths.

[0113] In this embodiment, the temporary profile construction module 401 acquires multi-user behavior data, social relationship data, and environmental data, and constructs temporary user profiles and group temporary profiles. The initial recommended path generation module 402 filters a set of attractions based on the group profile, constructs an attraction association graph, and generates an initial recommended path. The final recommended path generation module 403 acquires multi-user current behavior data, constructs a current user profile, compares the current user profile with the temporary user profile, outputs deviation data, triggers a path update request, and generates a final recommended path. By generating an initial recommended path using the temporary user profile and group temporary profile, constructing a current user profile based on multi-user current behavior data, and comparing the current user profile with the temporary user profile to generate a final recommended path, the system achieves path recommendation based on the real-time behavior of multiple users, improving the lack of coordination in group itineraries and the overall tourism experience.

[0114] Regarding the system in the above embodiments, the specific ways in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0115] For the system embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0116] Accordingly, this application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; and, when the one or more programs are executed by the one or more processors, causing the one or more processors to implement the travel itinerary recommendation method as described above. Figure 6 The diagram shown is a hardware structure diagram of any device with data processing capabilities that provides a travel itinerary recommendation system according to an embodiment of the present invention. Except for... Figure 6 In addition to the processor, memory, and network interface shown, any data processing device in the embodiment may also include other hardware depending on the actual function of the data processing device, which will not be described in detail here.

[0117] Accordingly, this application also provides a computer-readable storage medium storing computer instructions thereon, which, when executed by a processor, implement the travel itinerary recommendation method described above. The computer-readable storage medium can be an internal storage unit of any data-processing device as described in any of the foregoing embodiments, such as a hard disk or memory. The computer-readable storage medium can also be an external storage device, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc., equipped on the device. Furthermore, the computer-readable storage medium can include both internal storage units of any data-processing device and external storage devices. The computer-readable storage medium is used to store the computer program and other programs and data required by the data-processing device, and can also be used to temporarily store data that has been output or will be output.

[0118] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0119] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method for recommending travel itineraries, characterized in that, Includes the following steps: Collect multi-user behavior data, social relationship data, and environmental data respectively, and construct temporary user profiles and group profiles; Based on the group profile, a set of attractions is selected, an attraction relationship graph is constructed, and an initial recommended path is generated; Acquire current behavior data of multiple users, construct current user profiles, compare current user profiles with temporary user profiles, output deviation data, trigger path update requests, and generate final recommended paths.

2. The travel itinerary recommendation method as described in claim 1, characterized in that, In the steps of acquiring multi-user behavior data, social relationship data, and environmental data, and constructing temporary user profiles and group profiles: Acquire multi-user behavior data, social relationship data, and environmental data respectively, preprocess the data, and output the processed data; Extract user interest tags from the processed data to construct a temporary user profile; Based on the processed data, user groups are identified, user group characteristics are extracted, temporary group profiles are constructed, and group-level interest tags are generated.

3. The travel itinerary recommendation method as described in claim 2, characterized in that, In the step of extracting user interest tags from the processed data and constructing a temporary user profile: Behavioral features are extracted as user interest tags, and social and environmental features are integrated to construct temporary user profiles.

4. The travel itinerary recommendation method as described in claim 2, characterized in that, In the steps of identifying user groups based on the processed data, extracting user group features, constructing temporary group profiles, and generating group-level interest tags: Based on team ID identification of peer users, individual profiles within the group are merged to construct a temporary group profile and generate group-level interest tags.

5. The travel itinerary recommendation method as described in claim 2, characterized in that, In the steps of filtering a set of attractions based on group profiles, constructing an attraction relationship graph, and generating an initial recommended route: Construct a site association graph by collecting site collections; Generate a route based on the attraction association map and output the initial recommended route data.

6. The travel itinerary recommendation method as described in claim 5, characterized in that, Before the step of constructing a site association graph from a set of sites: A filtering threshold is set, and based on group-level interest tags, a set of attractions with a matching degree higher than the filtering threshold is selected from the attraction database.

7. The travel itinerary recommendation method as described in claim 5, characterized in that, In the steps of constructing a scenic spot association graph from a collection of scenic spots: Using attractions as nodes and edge weights equal to the travel time between attractions, we obtain an attraction association graph.

8. The travel itinerary recommendation method as described in claim 1, characterized in that, In the steps of acquiring current behavior data of multiple users, constructing current user profiles, comparing current user profiles with temporary user profiles, outputting deviation data, triggering path update requests, and generating final recommended paths: Real-time collection of user behavior data, generation of current user profile, comparison of current user profile and temporary user profile, calculation of deviation, and output of deviation data; Determine deviations, perform local and global optimizations on the path, and generate the final recommended path.

9. The travel itinerary recommendation method as described in claim 8, characterized in that, In the steps of determining deviations, performing local and global path optimization, and generating the final recommended path: If a single attraction deviates from the path, the route will be locally optimized. If two or more attractions deviate from each other, the path will be optimized globally.

10. A travel itinerary recommendation system, applied to the travel itinerary recommendation method as described in claim 1, characterized in that, It includes a temporary profile building module, an initial recommendation path generation module, and a final recommendation path generation module; among which: The temporary profile building module is used to acquire multi-user behavior data, social relationship data, and environmental data respectively, and to build temporary user profiles and group profiles. The initial recommended path generation module is used to filter the set of attractions based on the group profile, construct the attraction association graph, and generate the initial recommended path; The final recommended path generation module is used to obtain current behavior data of multiple users, construct the current user profile, compare the current user profile with the temporary user profile, output deviation data, trigger a path update request, and generate the final recommended path.

Images