Tourist attraction recommendation method, system, device and storage medium based on big data

Abstract

The present disclosure provides a tourist attraction recommendation method and system, device, and storage medium based on big data, which belongs to the field of data processing technology. The method includes: determining a first user matching a target user on an intelligent recommendation platform based on the target user's personal information; generating a first recommendation scheme for the target user based on the target user's historical behavior information and the first user's historical behavior information; optimizing the target user's first recommendation scheme based on spatiotemporal information to obtain a target recommendation scheme; wherein the spatiotemporal information includes time information and space information. The present disclosure can improve the accuracy of the recommendation scheme and enhance the user experience.

Description

Technical Field

[0001] The present disclosure belongs to the field of data processing technology, and more specifically, to a tourist attraction recommendation method, system, device, and storage medium based on big data. Background Art

[0002] With the rapid development of the tourism industry and the diversification of tourist needs, traditional travel planning methods are no longer able to meet the personalized needs of modern tourists. Traditional travel planning often relies on manually formulated itineraries and attraction recommendations, lacking precise recommendations tailored to tourists' individual needs, resulting in a poor user experience. Summary of the Invention

[0003] The purpose of this disclosure is to provide a tourist attraction recommendation method and system, device, and storage medium based on big data to improve the accuracy of recommendation solutions and enhance the user experience.

[0004] In a first aspect of the embodiments of the present disclosure, a method for recommending tourist attractions based on big data is provided, which is applied to an intelligent recommendation platform and includes:

[0005] Determining a first user on the intelligent recommendation platform who matches the target user based on the target user's personal information;

[0006] generating a first recommendation scheme for the target user based on the historical behavior information of the target user and the historical behavior information of the first user;

[0007] The first recommendation scheme for the target user is optimized based on the spatiotemporal information to obtain a target recommendation scheme; wherein the spatiotemporal information includes time information and space information.

[0008] A second aspect of the present disclosure provides a tourist attraction recommendation system based on big data, which is applied to an intelligent recommendation platform and includes:

[0009] A matching module, configured to determine a first user on the intelligent recommendation platform who matches the target user based on the target user's personal information;

[0010] A first determination module, configured to generate a first recommendation scheme for the target user based on the historical behavior information of the target user and the historical behavior information of the first user;

[0011] The solution recommendation module is used to optimize the first recommended solution of the target user based on the spatiotemporal information to obtain a target recommended solution; wherein the spatiotemporal information includes time information and space information.

[0012] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of the above-mentioned method for recommending tourist attractions based on big data when executing the computer program.

[0013] According to a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned method for recommending tourist attractions based on big data are implemented.

[0014] The beneficial effects of the tourist attraction recommendation method, system, device, and storage medium based on big data provided by the embodiments of the present disclosure are:

[0015] On the one hand, this disclosure fully utilizes the user data resources on the intelligent recommendation platform by accurately matching target users with similar users (primary users). Based on a comprehensive analysis of the personal information and historical behavior of both the target and primary users, a primary recommendation plan is generated that is more closely aligned with the target user's actual needs and preferences. This personalized recommendation approach not only improves recommendation accuracy but also enhances user experience satisfaction, making it easier for users to find travel plans that suit them.

[0016] Furthermore, this disclosure further incorporates spatiotemporal information to optimize the first recommended solution, resulting in a more practical and reasonable target recommendation. This consideration of spatiotemporal information allows the recommended solution to better adapt to the target user's travel time and location, avoiding itinerary issues caused by time conflicts or geographical inconvenience. This optimization not only enhances the practicality of the recommended solution but also improves the user's travel efficiency and the quality of their travel experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0018] Figure 1 A flowchart of a method for recommending tourist attractions based on big data according to an embodiment of the present disclosure;

[0019] Figure 2 This is a structural block diagram of a tourist attraction recommendation system based on big data provided by an embodiment of the present disclosure;

[0020] Figure 3A schematic block diagram of an electronic device provided in one embodiment of the present disclosure. DETAILED DESCRIPTION

[0021] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present disclosure. However, it will be apparent to those skilled in the art that the present disclosure may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present disclosure with unnecessary detail.

[0022] In order to make the purpose, technical solutions and advantages of the present disclosure more clear, specific embodiments will be described below with reference to the accompanying drawings.

[0023] Please refer to Figure 1 , Figure 1 This is a flow chart of a method for recommending tourist attractions based on big data according to an embodiment of the present disclosure. The method is applied to an intelligent recommendation platform and includes:

[0024] S101: Determine the first user who matches the target user on the intelligent recommendation platform based on the target user's personal information.

[0025] In this embodiment, the intelligent recommendation platform is a software system that utilizes artificial intelligence algorithms and big data technologies. It collects, stores, and analyzes various user data, including but not limited to basic user information, behavioral data (such as browsing history and purchasing behavior), and feedback data. Based on this data, it then provides users with personalized travel planning recommendations.

[0026] A target user is a specific individual targeted by the intelligent recommendation platform. This individual has unique personal information, such as age, gender, interests, spending habits, and geographic location. The platform collects and analyzes target user information to better understand their needs and preferences, thereby finding other users or recommended content that matches them.

[0027] First users are other users on the intelligent recommendation platform who have a high degree of compatibility with the target user. This matching is usually based on a series of matching rules and algorithms set by the platform. These rules involve multiple dimensions of user personal information, such as similarity of interests and hobbies, consistency of consumption behavior, and relevance of social circles.

[0028] Determining the first user who matches the target user on the intelligent recommendation platform based on the target user's personal information includes:

[0029] Build a user profile for each user based on the user information on the intelligent recommendation platform, and calculate the similarity between the target user profile and the platform user profile based on a preset algorithm;

[0030] The first user is determined based on the similarity between the target user profile and the user profiles of the platform users.

[0031] In this embodiment, a user profile is constructed for each user based on their information on the intelligent recommendation platform. This profile includes various user information, such as basic attributes (age, gender, and geographic location), interests and hobbies (favorite movie genres, music styles, and sports), and consumer behavior (purchase frequency, product categories, and spending amounts). The similarity between the target user profile and other user profiles is calculated using algorithms such as cosine similarity, Euclidean distance, and the Pearson correlation coefficient, and the user with the highest similarity is selected as the primary user. For example, on the intelligent recommendation platform, if the target user and another user share many common tags in the interests dimension (e.g., both like science fiction and historical documentaries), their similarity in this dimension will be relatively high.

[0032] Determining the first user who matches the target user on the intelligent recommendation platform based on the target user's personal information also includes:

[0033] The social relationships of users are determined based on the information of users on the intelligent recommendation platform, and the first user matching the target user on the intelligent recommendation platform is determined based on the social relationships.

[0034] In this embodiment, the social relationship is determined based on the friend list information and / or the watch list information.

[0035] Smart recommendation platforms can identify users' social relationships based on their information. For example, smart recommendation platforms typically record users' friend lists. Beyond explicitly added friendships, potential friendships can also be inferred through interactions. For example, if user C frequently likes and comments on user D's travel notes, and user D also interacts with user C's related content, while they may not have officially added each other as friends, it can be assumed that they have a potential friendship based on learning, communication, and interaction.

[0036] When matching a first user based on the target user's friend list, other users who share a first number of mutual friends with the target user are identified and designated as the first user. This first number is proportional to the target user's total number of friends. For example, if the target user has S total friends, the first number is kS, where k∈(0.2, 1). For example, if target user E has friends F, G, H, etc., the platform will analyze the friend lists of other users and discover that user I also has friends F and G. Therefore, user I has a high number of mutual friends with target user E.

[0037] S102: Generate a first recommendation scheme for the target user based on the historical behavior information of the target user and the historical behavior information of the first user.

[0038] In this embodiment, historical behavior information refers to a user's travel-related behavior and activity records over a period of time, including but not limited to travel destination selection, accommodation reservations, transportation options, scenic spot visits, dining expenses, and shopping history. Recommendations are suggestions or plans generated based on user historical behavior information and algorithmic analysis, designed to meet the user's travel needs. Recommendations may include recommended travel destinations, accommodation options, activity arrangements, transportation methods, and more. In this embodiment, both the target user's historical behavior information and the first user's historical behavior information may include travel history, travel activities, accommodation preferences, and other information. Travel history includes the target user's past travel destinations, such as popular tourist cities like Beijing, Shanghai, and Paris, or areas with less popular attractions. The time span of travel is also crucial, determining whether the user prefers to travel during holidays or utilizes off-peak travel, such as during annual leave. For example, some users consistently travel during long holidays like National Day and Spring Festival, suggesting a preference for enjoying the travel atmosphere when everyone has free time. Meanwhile, some users prefer to travel during non-holiday periods like May or October, perhaps to avoid peak crowds and enjoy a quieter travel experience.

[0039] The activities one has participated in at a travel destination can also reflect their preferences. For example, when traveling to the seaside, do you prefer to simply sunbathe and play in the water on the beach, or are you keen on participating in water sports such as surfing and diving? When traveling in the mountains, do you prefer to hike up the mountain to enjoy the natural scenery, or do you choose to take a cable car or other convenient means of sightseeing? Accommodation preferences: that is, do you prefer to stay in a star-rated hotel and enjoy comfortable service, or do you prefer to choose a unique homestay to experience the local customs. For example, some users choose to stay in a local homestay every time they travel, perhaps because they value the homestay's ability to provide an experience closer to local life, allowing them to communicate with the homestay owner and other guests and gain a deeper understanding of the local culture. However, some users will choose star-rated hotels in large cities to ensure service quality and convenience, and choose homestays in places with beautiful scenery but relatively simple supporting facilities, depending on the situation of the travel destination.

[0040] In this embodiment, a first recommendation scheme for the target user is generated based on the target user's historical behavior information and the first user's historical behavior information. The following steps are performed: The target user and the first user's historical behavior information is collected, and the collected data is cleaned to remove invalid or abnormal data. The target user and the first user's explicit preferences for travel destinations, activities, accommodation, etc., such as high ratings and positive reviews, are analyzed. Implicit feedback data, such as browsing behavior, dwell time, and click-through rate, from the target user and the first user is analyzed to uncover their potential travel preferences. Similarity calculation methods, such as cosine similarity and Pearson correlation, can be used to calculate the similarity between the target user and the first user based on their historical behavior information. Based on the calculated similarity, a set of first users most similar to the target user is selected. Based on this selected set of similar users, the travel destinations, activities, accommodation, etc., commonly preferred by these users, are analyzed as a candidate set for recommendation to the target user. Combined with the target user's historical behavior information and preferences, the candidate set is further screened and optimized to generate a personalized recommendation scheme for the target user.

[0041] S103: Optimizing the first recommendation scheme for the target user based on the spatiotemporal information to obtain a target recommendation scheme; wherein the spatiotemporal information includes time information and space information.

[0042] In this embodiment, spatiotemporal information includes time information and spatial information. In travel recommendations, spatiotemporal information may include the user's travel time (e.g., season, holidays), travel locations (e.g., cities, scenic spots), and the spatial relationships between locations (e.g., distance, transportation methods).

[0043] The first recommendation is a preliminary proposal generated based on the user's historical behavior information and other relevant data (such as user preferences and budget). This proposal includes a series of travel destinations, activities, accommodations, and other recommendations. The targeted recommendation is the final, optimized proposal. This proposal not only considers the target user's historical behavior information and preferences, but also incorporates temporal and spatial information to provide more precise and practical travel recommendations. Targeted recommendations are generally more aligned with user expectations and needs, helping to increase user satisfaction and loyalty.

[0044] In this embodiment, spatiotemporal data, including the target user's travel plans, destinations, and itinerary, is collected. The target user's travel plans are analyzed, taking into account the impact of different seasons and holidays on travel choices. The recommended plan adjusts the activities and order of attractions, etc., based on time factors. Geographical information related to the tourist destinations, the distribution of attractions, and traffic conditions are obtained. The geographical relationships between tourist destinations are analyzed, and the distances and transportation methods between attractions are calculated. The target user's travel methods and preferences are considered to optimize itinerary routes and transportation arrangements. The target user's travel plans, destinations, itinerary, and other spatiotemporal data can be information the target user has entered into the intelligent recommendation platform. The intelligent recommendation platform inputs this user-entered information and the first recommended plan into a data analysis model (convolutional neural network model) to generate the target recommended plan. For example, the first recommended plan is to travel to Beijing during the October 1st holiday, visiting destination A on the first day, destination B on the second day, and destination C on the third day, using the subway as the means of transportation. However, the target user has entered information on the platform, planning to travel to a less crowded destination with an open view on October 1st. Based on the target user's input, the intelligent recommendation platform determines that Beijing is a popular destination during the October 1st holiday, resulting in a high population density and limited viewing areas. Therefore, the first recommendation is changed to a trip to Ulanqab during the October 1st holiday, with a destination aa on the first day, bb on the second day, and cc on the third day. Given the distances between the various destinations, a long-distance bus is chosen as the preferred mode of transportation.

[0045] Based on the results of spatiotemporal analysis, filter out travel destinations and activities that meet the needs and preferences of target users.

[0046] The selected recommendations are sorted, prioritizing attractions and activities that are close, convenient, and highly rated. Recommendations are personalized based on the target user's personal preferences and specific needs. For example, local restaurants may be recommended based on the user's food preferences, and suitable entertainment activities may be recommended based on the user's leisure needs. The platform integrates real-time information such as weather conditions and visitor flow at scenic spots through access to weather forecast platforms and online scenic spot monitoring systems. Recommendations are dynamically adjusted based on this real-time information to ensure the accuracy and practicality of the recommendations. For example, if the platform detects that heavy rainfall is expected at a recommended location during the target user's planned travel time, the recommended destination is modified. If the platform detects that the target scenic spot at a recommended location is fully booked during the planned travel time, the recommended destination is modified. The optimized recommendations are comprehensively evaluated to ensure that they meet the target user's needs in terms of time, space, and user preferences. This comprehensive evaluation can be performed by matching the recommended destination, scenic spot, route, and transportation with the user's needs and assigning a corresponding matching score. If the score exceeds a preset score (set by the platform), the user's needs are considered met. For example, if the destination matches the user's needs with a score of 70, the scenic spot matches the user's needs with a score of 90, the route matches the user's needs with a score of 80, and the transportation matches the user's needs with a score of 90, and the average score is 82.5 (greater than the preset score of 80), then the recommended solution meets the needs of the target user.

[0047] Based on the evaluation results, a final target recommendation plan is generated. This plan should include a detailed itinerary, attraction descriptions, transportation options, accommodation suggestions, and more. Recommendations should balance popular attractions (which users may be interested in but haven't visited yet) with lesser-known attractions (which meet users' unique preferences) to increase recommendation diversity. Furthermore, using the Maps API, an optimal itinerary is generated based on the recommended attraction list, taking into account time efficiency, accessibility, and a logical order between attractions (e.g., avoiding duplicate routes and sorting by points of interest).

[0048] As can be seen from the above, this disclosure fully utilizes the user data resources on the intelligent recommendation platform by accurately matching target users with similar users (first users). Based on a comprehensive analysis of the personal information and historical behavior of both the target user and the first user, a first recommendation plan is generated that is more closely aligned with the target user's actual needs and preferences. This personalized recommendation approach not only improves recommendation accuracy but also enhances user experience satisfaction, making it easier for users to find travel plans that suit them.

[0049] Furthermore, this disclosure further incorporates spatiotemporal information to optimize the first recommended solution, resulting in a more practical and reasonable target recommendation. This consideration of spatiotemporal information allows the recommended solution to better adapt to the target user's travel time and location, avoiding itinerary issues caused by time conflicts or geographical inconvenience. This optimization not only enhances the practicality of the recommended solution but also improves the user's travel efficiency and the quality of their travel experience.

[0050] In one embodiment of the present disclosure, optimizing a first recommendation scheme for a target user based on spatiotemporal information to obtain a target recommendation scheme includes:

[0051] Based on the spatiotemporal information and the input information of the target user on the intelligent recommendation platform, the first recommendation plan of the target user is optimized to obtain the target recommendation plan.

[0052] The input information includes text information or voice information;

[0053] If the input information is text information, the first recommendation scheme for the target user is optimized based on the spatiotemporal information and the text information to obtain a target recommendation scheme;

[0054] If the input information is voice information, the voice features are determined based on the voice information, and the first recommendation scheme for the target user is optimized based on the spatiotemporal information and the voice features to obtain a target recommendation scheme.

[0055] In this embodiment, input information refers to additional information provided by the target user on the intelligent recommendation platform to further optimize the recommendation solution. This information can be text (such as specific requirements and preferences entered by the user) or voice (such as requirements and questions entered by the user via voice).

[0056] Input information is distinct from the target user's personal information and historical behavior information. Input information refers to the information a target user enters into the smart recommendation platform when they have travel needs; personal information refers to the information they fill out when registering for the smart recommendation platform; and historical behavior information refers to information the smart recommendation platform obtains from the target user's other connected apps, such as Taobao, 12306, Tongcheng Travel, and Xiaohongshu. The smart recommendation platform can obtain information from the target user's other connected apps through various means, including user authorization and cross-app ID technology. For example, when a target user uses the smart recommendation platform, the platform will request certain permissions, such as accessing their address book, location information, and app list. Once the target user clicks the authorization button, they consent to the platform obtaining their relevant information. Cross-app ID technology is used to identify the same user across different apps. Through this technology, the smart recommendation platform can track the target user's behavior across different apps, thereby collecting more comprehensive historical behavior information. However, the use of cross-app ID technology must comply with relevant laws, regulations, and privacy policies to ensure the legality and security of user data.

[0057] Text information refers to information entered by users on intelligent recommendation platforms via keyboards or other text input devices. This information typically includes specific user needs, preferences, and constraints. Voice information refers to information entered by users on intelligent recommendation platforms via voice input devices (such as microphones). This information requires voice recognition and processing to extract the user's intent and needs. Voice features are distinguishing characteristics extracted from voice information, such as language type, speaking rate, intonation, and timbre. These features help the system more accurately understand the user's voice input and are used to optimize recommendations.

[0058] The targeted recommendation is the final, optimized travel plan. This plan not only considers the target user's personal information and historical behavior, but also incorporates temporal and spatial information, as well as real-time user input (whether text or voice) on the intelligent recommendation platform. Targeted recommendations are generally more aligned with user expectations and needs, helping to increase user satisfaction and loyalty.

[0059] Determining speech features based on speech information includes: recognizing the speech information to obtain first audio information, and extracting features from the first audio information to obtain speech features. Speech features include basic physical features, linguistic features, and speech features. Basic physical features include pitch, volume, and timbre. Linguistic features include phonemes, syllables, intonation, rhythm, and continuity. Speech features include pronunciation habits and emotions.

[0060] The first recommendation solution for the target user is optimized based on spatiotemporal information and voice features to obtain a target recommendation solution, including:

[0061] Obtaining a first location, travel time, and voice characteristics of a target user; determining weather characteristics, traffic characteristics, and scenic spot information of a second location based on the first location and travel time of the target user; the second location being the location of the travel destination;

[0062] Based on the weather characteristics, traffic characteristics, scenic spot information and voice characteristics of the second location, the voice guide, tourist attraction sequence and transportation mode in the first recommendation plan are optimized to obtain a target recommendation plan.

[0063] In this embodiment, the first location is the target user's current location. The target user's first location and travel time are determined based on spatiotemporal information, and the audio guide in the first recommended solution is determined based on voice features. The voice features also include the target user's language preference and accent. Determining the audio guide in the first recommended solution based on the voice features includes: determining the audio guide language based on the target user's language preference; and adjusting the speed and intonation of the audio guide based on the target user's accent.

[0064] For example, a smart travel recommendation platform receives a request from a target user, Mr. Zhang, for travel planning in Beijing within the next three days and requests a personalized travel recommendation. The system first generates a preliminary first recommendation based on Mr. Zhang's needs. To better meet Mr. Zhang's needs, the system can further optimize this first recommendation based on spatiotemporal information and voice features.

[0065] First location: Mr. Zhang's current city (e.g., Shanghai). Travel time: Mr. Zhang plans to travel to Beijing within the next three days, including the specific departure date and time. Voice characteristics: Through Mr. Zhang's voice input when interacting with the system, the system acquires Mr. Zhang's pronunciation habits, speaking speed, and other voice characteristics.

[0066] Second location: Beijing, Mr. Zhang's travel destination. Weather characteristics: The system queries real-time weather data to obtain Beijing's weather conditions for the next three days, including temperature, humidity, wind speed, etc. Traffic characteristics: The system analyzes Beijing's traffic conditions, including congestion on major roads and public transportation operating hours. Attraction information: The system compiles information about Beijing's major tourist attractions, including opening hours, ticket prices, and visitor reviews.

[0067] Optimize the first recommended solution:

[0068] Audio Guide: Based on Mr. Zhang's voice characteristics, the system optimized the audio guide in the first recommended solution. For example, the speech speed and intonation were adjusted to better suit Mr. Zhang's listening habits. The system also offers multiple language options to suit Mr. Zhang's language preferences.

[0069] Tourist attraction order: The system adjusts the order of tourist attractions in the first recommended plan based on Beijing's weather and traffic conditions. For example, outdoor attractions are scheduled for sunny and warm days, while indoor attractions are scheduled for days with a potential for rain or cooler temperatures. The system also considers transportation connections between attractions to minimize Mr. Zhang's travel time.

[0070] Transportation: The system recommends the most suitable transportation method for Mr. Zhang based on his travel time and Beijing's traffic conditions. For example, during peak hours, it recommends taking the subway or bus to avoid traffic congestion; at night or in remote areas, it recommends taking a taxi or booking a private car service.

[0071] After the above optimization, the system generated a more personalized, accurate, and easy-to-understand target recommendation plan for Mr. Zhang. The plan includes:

[0072] Detailed introduction to tourist attractions and the order of touring; personalized audio guide service; suitable transportation suggestions; real-time weather and traffic condition reminders; other travel suggestions, such as dining, accommodation, etc.

[0073] As can be seen from the above, this embodiment, by comprehensively considering spatiotemporal information and the target user's input information (whether text or voice) on the intelligent recommendation platform, can more accurately understand and meet the target user's personalized needs. For voice information, voice features are extracted to further optimize the recommendation scheme, making the recommendations more tailored to the user's listening habits and preferences, thereby improving the intelligence level of travel planning recommendations and the user experience.

[0074] In one embodiment of the present disclosure, determining a first user matching the target user on the intelligent recommendation platform based on the target user's personal information includes:

[0075] Based on the clustering algorithm, the registered users on the intelligent recommendation platform are grouped to obtain multiple user groups;

[0076] Determine the target user group based on the similarity between the personal information characteristics of each user group and the personal information characteristics of the target user;

[0077] At least one user is selected from the target user group as the first user.

[0078] In this embodiment, registered users refer to users who have registered on the intelligent recommendation platform and provided personal information in order to receive recommendation services. Such user information generally includes but is not limited to age, gender, geographic location, interests, and hobbies.

[0079] A clustering algorithm is a data mining technique used to divide a set of data objects into several groups or clusters, so that data objects within the same group have a high degree of similarity, while data objects between different groups have a low degree of similarity. In this embodiment, a clustering algorithm is used to group registered users.

[0080] A user group is a set of users obtained through a clustering algorithm, and the users in each user group have similar personal information characteristics.

[0081] Selecting at least one user from the target user group as the first user includes:

[0082] The target weights of the users are determined according to the weights of the different user information, and at least one user is selected as the first user according to the target weights.

[0083] In this embodiment, the system first collects personal information from users registered on the platform. This information includes, but is not limited to, age, gender, location, interests, and historical behavior. This collected personal information is then preprocessed, such as through deduplication, cleaning, and formatting, to ensure data accuracy and consistency. A clustering algorithm (such as K-means clustering or hierarchical clustering) is then applied to group the preprocessed registered users. Clustering algorithms can divide users into multiple groups based on the similarity of their personal information features. Users within each group share similar personal information features and behavioral patterns.

[0084] For the target user, extract their personal information features. Calculate the similarity between the target user's personal information features and the central features of each user group (using metrics such as cosine similarity and Euclidean distance). Then sort by similarity and select the user group with the highest similarity as the target user group. Within the target user group, select at least one user as the first user based on their characteristics. Leveraging information such as the first user's preferences and behavior patterns, provide personalized recommendations for the target user.

[0085] As can be seen from the above, the disclosed embodiments effectively group registered users on the intelligent recommendation platform through a clustering algorithm and accurately determine a target user group that matches the target user based on the similarity of their personal information characteristics. At least one user from this group is then selected as the first user. These first users have a high degree of similarity with the target user and can more accurately reflect the target user's preferences and needs, thereby greatly improving the personalization and accuracy of intelligent recommendations and enhancing the user experience.

[0086] In one embodiment of the present disclosure, generating a first recommendation scheme for a target user based on historical behavior information of the target user and historical behavior information of a first user includes:

[0087] Determine similarities and differences between the target user and the first user's historical behavior information;

[0088] The historical behavior information of the first user is weighted based on the similarities and differences to generate a first recommendation solution for the target user.

[0089] In this embodiment, commonalities refer to the portion of the target user's historical behavior information with the first user where the similarity exceeds a first threshold; differences refer to the portion of the target user's historical behavior information with the first user where the similarity is less than a second threshold; the first threshold is greater than the second threshold. The first and second thresholds are preset by the intelligent recommendation platform. The first threshold can be 0.7, and the second threshold can be 0.5.

[0090] In this embodiment, the system first analyzes the historical behavior information of the target user and the first user. Then, the system determines the similarities and differences in the historical behavior information of the target user and the first user. The similarities reflect the common interests or needs of the target user and the first user. The differences reflect the personalized differences between the target user and the first user in certain aspects. Finally, the system weights the historical behavior information of the first user based on the similarities and differences. The system assigns higher weights to the similarities because these parts can better reflect the real needs of the target user. For the differences, the system will weight them to different degrees according to the specific situation to balance the needs of personalized recommendations and common recommendations. Based on the weighted historical behavior information of the first user, a first recommendation plan for the target user is generated.

[0091] It can be concluded from the above that the target recommendation scheme generated by this embodiment not only includes the content that the target user is interested in, but also takes into account the personalized differences between the target user and the first user, thereby improving the accuracy and relevance of the recommendation.

[0092] In one embodiment of the present disclosure, weighting the historical behavior information of the first user based on the similarities and differences to generate a first recommendation scheme for the target user includes:

[0093] Determining a weight corresponding to each item in the historical behavior information of the first user based on the similarities and differences;

[0094] The historical behavior information of the first user is weighted based on the weight corresponding to each item to generate a first recommendation plan for the target user.

[0095] In this embodiment, weights are calculated for each item in the first user's historical behavior information (e.g., attraction style, location, transportation, etc.) based on similarities and differences. Items with more similarities can be assigned higher weights, as they better align with the target user's preferences. For items with significant differences, weights can be adjusted based on actual circumstances. If these differences are deemed to have no negative impact on recommendations or to provide new points of interest for the target user, they can be given a certain weight. If these differences are deemed to conflict with the target user's preferences, their weights can be reduced or ignored. Based on the calculated weights, the first user's historical behavior information is weighted. Based on this weighted historical behavior information, a first recommendation for the target user is generated.

[0096] As can be seen from the above, this embodiment calculates the weight of each item in the first user's historical behavior information and fully considers the similarities and differences between the target user and the first user, making the recommendation plan more personalized to the target user's needs. Furthermore, this embodiment can flexibly adjust the weight of the difference items, avoiding inaccurate recommendations due to excessive differences while providing the target user with the opportunity to explore new points of interest, thereby improving recommendation accuracy and user satisfaction.

[0097] In one embodiment of the present disclosure, determining a weight corresponding to each item in the historical behavior information of the first user based on the similarities and differences includes:

[0098] determining a difference threshold based on the difference points;

[0099] If the difference value of the item corresponding to the difference point is greater than the difference threshold, the item corresponding to the difference point is assigned a weight in the first range or the item corresponding to the difference point is excluded from the first recommendation scheme;

[0100] If the difference value of the item corresponding to the difference point is less than or equal to the difference threshold, the item corresponding to the difference point is assigned a weight in the second range;

[0101] The weight of the first range is smaller than the weight of the second range.

[0102] In this embodiment, the difference value is a numerical value that quantifies the size or degree of a difference. It reflects the degree of difference between the target user and the first user's preferences for a specific item. The difference threshold is a pre-set numerical limit used to determine the degree of influence of a difference point on the item weight assignment. A difference point with a difference value less than this threshold is considered a significant difference; otherwise, it is considered a minor difference. The difference threshold in this embodiment is a threshold preset by the intelligent recommendation platform. The setting of the difference threshold is related to the second threshold, and the difference threshold is less than the second threshold. For example, the difference threshold can be 50% of the second threshold. When the difference value of a difference point is greater than the difference threshold, a weight range is assigned to the item corresponding to the difference point. This range of weights is generally lower, indicating that the item differs significantly from the target user's preferences and therefore has lower importance in the recommendation scheme. When the difference value of a difference point is less than or equal to the difference threshold, a weight range is assigned to the item corresponding to the difference point. This range of weights is generally higher, indicating that the item is relatively close to the target user's preferences and therefore has higher importance in the recommendation scheme.

[0103] As can be seen from the above, this embodiment, by setting a difference threshold and precisely adjusting item weights, can effectively distinguish between the preferences of the target user and the first user on different items. Items with significant differences are weighted down or excluded to avoid irrelevant recommendations; items with smaller differences are weighted appropriately to ensure personalized recommendations. This embodiment can significantly improve recommendation accuracy and user satisfaction, making recommendations more tailored to the actual needs of the target user.

[0104] In one embodiment of the present disclosure, determining the weight corresponding to each item in the historical behavior information of the first user based on the similarities and differences further includes:

[0105] Determine a similarity threshold based on the commonalities;

[0106] If the similarity of the items corresponding to the same point is greater than the similarity threshold, the items corresponding to the same point are directly added to the first recommendation scheme;

[0107] If the similarity of the items corresponding to the same point is less than or equal to the similarity threshold, the items corresponding to the same point are assigned a weight in the third range.

[0108] In this embodiment, the system sets one or more difference thresholds based on the differences in historical behavioral information between the target user and the first user. These thresholds are used to determine which items have significant differences and which differences are acceptable or small. The system then compares the difference value of each item with the difference threshold. If the difference value of an item is greater than the set threshold, it means that the item may not be very relevant to the target user or there is a large difference in preference, so the system will reduce the weight of the item or even choose to exclude it from the recommendation scheme completely. On the contrary, if the difference value of an item is less than or equal to the threshold, it means that the item is closer to the target user's preference, and the system will moderately increase the weight of the item. In this way, the system can effectively distinguish the preference differences between the target user and the first user on different items, and thus more accurately determine which items should be recommended to the target user.

[0109] As can be seen above, this embodiment further refines the item weighting strategy by setting a similarity threshold. Highly similar items (exceeding the threshold) are directly added to the recommendation scheme to ensure that core preferences are met. Items with lower similarity are assigned appropriate weights to balance personalization and breadth. This embodiment not only improves the accuracy of recommendations, but also enhances recommendation diversity and user satisfaction.

[0110] Corresponding to the tourist attraction recommendation method based on big data in the above embodiment, Figure 2 This is a structural block diagram of a tourist attraction recommendation system based on big data provided by an embodiment of the present disclosure. For ease of explanation, only the parts related to the embodiment of the present disclosure are shown. Figure 2 The tourist attraction recommendation system 20 based on big data is applied to the intelligent recommendation platform, including: a matching module 21, a first determination module 22 and a solution recommendation module 23.

[0111] The matching module 21 is configured to determine a first user matching the target user on the intelligent recommendation platform based on the target user's personal information;

[0112] A first determination module 22 is configured to generate a first recommendation scheme for the target user based on the historical behavior information of the target user and the historical behavior information of the first user;

[0113] The solution recommendation module 23 is configured to optimize the first recommended solution for the target user based on the spatiotemporal information to obtain a target recommended solution; wherein the spatiotemporal information includes time information and space information.

[0114] In one embodiment of the present disclosure, the solution recommendation module 23 is specifically configured to:

[0115] Based on the spatiotemporal information and the input information of the target user on the intelligent recommendation platform, the first recommendation plan of the target user is optimized to obtain the target recommendation plan.

[0116] In one embodiment of the present disclosure, the input information includes text information or voice information;

[0117] The solution recommendation module 23 is specifically used for:

[0118] If the input information is text information, the first recommendation scheme for the target user is optimized based on the spatiotemporal information and the text information to obtain a target recommendation scheme;

[0119] If the input information is voice information, the voice features are determined based on the voice information, and the first recommendation scheme for the target user is optimized based on the spatiotemporal information and the voice features to obtain a target recommendation scheme.

[0120] In one embodiment of the present disclosure, the matching module 21 is specifically configured to:

[0121] Based on the clustering algorithm, the registered users on the intelligent recommendation platform are grouped to obtain multiple user groups;

[0122] Determine the target user group based on the similarity between the personal information characteristics of each user group and the personal information characteristics of the target user;

[0123] At least one user is selected from the target user group as the first user.

[0124] In one embodiment of the present disclosure, the first determining module 22 is specifically configured to:

[0125] Determine similarities and differences between the target user and the first user's historical behavior information;

[0126] The historical behavior information of the first user is weighted based on the similarities and differences to generate a first recommendation solution for the target user.

[0127] In one embodiment of the present disclosure, the first determining module 22 is specifically configured to:

[0128] Determining a weight corresponding to each item in the historical behavior information of the first user based on the similarities and differences;

[0129] The historical behavior information of the first user is weighted based on the weight corresponding to each item to generate a first recommendation plan for the target user.

[0130] In one embodiment of the present disclosure, the first determining module 22 is specifically configured to:

[0131] determining a difference threshold based on the difference points;

[0132] If the difference value of the item corresponding to the difference point is greater than the difference threshold, the item corresponding to the difference point is assigned a weight in the first range or the item corresponding to the difference point is excluded from the first recommendation scheme;

[0133] If the difference value of the item corresponding to the difference point is less than or equal to the difference threshold, the item corresponding to the difference point is assigned a weight in the second range;

[0134] The weight of the first range is smaller than the weight of the second range.

[0135] See also Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided by an embodiment of the present disclosure. Figure 3 The electronic device 300 in the embodiment shown may include: one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memory 304 is used to store computer programs, which include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. The processor 301 is configured to call the program instructions to execute the functions of the modules / units in the above-mentioned device embodiments, such as Figure 2 The functions of modules 21 to 23 are shown.

[0136] It should be understood that in the embodiments of the present disclosure, the processor 301 may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0137] The input device 302 may include a touchpad, a fingerprint collection sensor (for collecting user fingerprint information and fingerprint direction information), a microphone, etc. The output device 303 may include a display (LCD, etc.), a speaker, etc.

[0138] The memory 304 may include a read-only memory and a random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include a non-volatile random access memory. For example, the memory 304 may also store device type information.

[0139] In a specific implementation, the processor 301, input device 302, and output device 303 described in the embodiments of the present disclosure can execute the implementation methods described in the first and second embodiments of the tourist attraction recommendation method based on big data provided by the embodiments of the present disclosure, and can also execute the implementation methods of the electronic device described in the embodiments of the present disclosure, which will not be repeated here.

[0140] In another embodiment of the present disclosure, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, all or part of the process of the method in the above embodiment is implemented. The computer program can also be used to instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of each of the above method embodiments are implemented. The computer program includes computer program code, which can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium.

[0141] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the aforementioned embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. Furthermore, the computer-readable storage medium can include both an internal storage unit of the electronic device and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or is about to be output.

[0142] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the composition and steps of each example according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0143] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the electronic devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0144] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces or units, or can be an electrical, mechanical or other form of connection.

[0145] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, i.e., they may be located in one place or distributed across multiple network units. Some or all of these units may be selected based on actual needs to achieve the objectives of the embodiments of the present disclosure.

[0146] In addition, the functional units in the various embodiments of the present disclosure may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0147] The above are only specific embodiments of the present disclosure, but the scope of protection of the present disclosure is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or replacements within the technical scope disclosed in this disclosure, and such modifications or replacements should be included in the scope of protection of the present disclosure. Therefore, the scope of protection of the present disclosure should be based on the scope of protection of the claims.

Claims

1. A tourist attraction recommendation method based on big data, characterized in that: Applied to intelligent recommendation platforms, including: Determining a first user on the intelligent recommendation platform that matches the target user based on the target user's personal information; Generating a first recommendation scheme for the target user based on the historical behavior information of the target user and the historical behavior information of the first user includes: Determine the commonalities and differences in the historical behavior information of the target user and the first user; the commonalities refer to the parts of the historical behavior information of the target user and the first user where the similarity is greater than a first threshold; the differences refer to the parts of the historical behavior information of the target user and the first user where the similarity is less than a second threshold; the first threshold is greater than the second threshold; Determine a weight corresponding to each item in the historical behavior information of the first user based on the common points and the differences; Weighting the historical behavior information of the first user based on the weight corresponding to each item to generate a first recommendation plan for the target user; The determining, based on the common points and the differences, a weight corresponding to each item in the historical behavior information of the first user includes: determining a difference threshold based on the difference point; If the difference value of the item corresponding to the difference point is greater than the difference threshold, assigning the item corresponding to the difference point a weight in a first range or excluding the item corresponding to the difference point from the first recommendation scheme; If the difference value of the item corresponding to the difference point is less than or equal to the difference threshold, assigning the item corresponding to the difference point a weight in the second range; The weight of the first range is less than the weight of the second range; Determine a similarity threshold based on the commonalities; If the similarity of the items corresponding to the same point is greater than the similarity threshold, the items corresponding to the same point are directly added to the first recommendation scheme; If the similarity of the items corresponding to the same point is less than or equal to the similarity threshold, the items corresponding to the same point are assigned a weight in the third range; The first recommendation scheme for the target user is optimized based on the spatiotemporal information to obtain a target recommendation scheme; wherein the spatiotemporal information includes time information and space information.

2. The tourist attraction recommendation method based on big data according to claim 1, characterized in that: The step of optimizing the first recommendation scheme for the target user based on the spatiotemporal information to obtain the target recommendation scheme includes: Based on the spatiotemporal information and the input information of the target user on the intelligent recommendation platform, the first recommendation scheme for the target user is optimized to obtain a target recommendation scheme.

3. The tourist attraction recommendation method based on big data as claimed in claim 2, characterized in that: The input information includes text information or voice information; The step of optimizing the first recommendation scheme for the target user based on the spatiotemporal information and the input information of the target user on the intelligent recommendation platform to obtain a target recommendation scheme includes: If the input information is text information, optimizing the first recommendation scheme for the target user based on the spatiotemporal information and the text information to obtain a target recommendation scheme; If the input information is voice information, voice features are determined based on the voice information, and the first recommendation scheme for the target user is optimized based on the spatiotemporal information and the voice features to obtain a target recommendation scheme.

4. The tourist attraction recommendation method based on big data according to claim 1, characterized in that: The determining, based on the personal information of the target user, a first user matching the target user on the intelligent recommendation platform comprises: Registered users on the intelligent recommendation platform are grouped based on a clustering algorithm to obtain multiple user groups; Determine the target user group based on the similarity between the personal information characteristics of each user group and the personal information characteristics of the target user; At least one user is selected from the target user group as a first user.

5. The tourist attraction recommendation system based on big data is characterized by: Applied to intelligent recommendation platforms, including: A matching module, configured to determine a first user on the intelligent recommendation platform that matches the target user based on the target user's personal information; A first determination module, configured to generate a first recommendation scheme for the target user based on the historical behavior information of the target user and the historical behavior information of the first user, comprises: Determine the commonalities and differences in the historical behavior information of the target user and the first user; the commonalities refer to the parts of the historical behavior information of the target user and the first user where the similarity is greater than a first threshold; the differences refer to the parts of the historical behavior information of the target user and the first user where the similarity is less than a second threshold; the first threshold is greater than the second threshold; Determine a weight corresponding to each item in the historical behavior information of the first user based on the common points and the differences; Weighting the historical behavior information of the first user based on the weight corresponding to each item to generate a first recommendation plan for the target user; The determining, based on the common points and the differences, a weight corresponding to each item in the historical behavior information of the first user includes: determining a difference threshold based on the difference point; If the difference value of the item corresponding to the difference point is greater than the difference threshold, assigning the item corresponding to the difference point a weight in a first range or excluding the item corresponding to the difference point from the first recommendation scheme; If the difference value of the item corresponding to the difference point is less than or equal to the difference threshold, assigning the item corresponding to the difference point a weight in the second range; The weight of the first range is less than the weight of the second range; Determine a similarity threshold based on the commonalities; If the similarity of the items corresponding to the same point is greater than the similarity threshold, the items corresponding to the same point are directly added to the first recommendation scheme; If the similarity of the items corresponding to the same point is less than or equal to the similarity threshold, the items corresponding to the same point are assigned a weight in the third range; The solution recommendation module is used to optimize the first recommended solution of the target user based on the spatiotemporal information to obtain a target recommended solution; wherein the spatiotemporal information includes time information and space information.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 4 are implemented.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

Images