Multi-factor fused interest point recommendation method and recommendation system
By integrating multi-factor interest recommendation methods, users' interest information is comprehensively collected, multiple factors are considered, and a similarity matrix is constructed using advanced algorithms to generate personalized recommendation lists. This solves the problem of insufficient recommendation results in the existing technology, and achieves higher recommendation accuracy and personalization.
Patent Information
- Application Number
- CN202510071083.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-16
- Publication Date
- 2025-05-13
AI Technical Summary
The existing interest point recommendation system usually only recommends based on the user's geographical location and historical behavior, ignoring the diversity, complexity of user interest points, as well as multi-dimensional information such as time and space, resulting in the recommendation results being insufficiently accurate and unable to meet the user's personalized needs.
A multi-factor interest recommendation method is adopted to comprehensively collect user interest information through the information collection unit, including user name, location, pictures and other data. In the analysis unit, a variety of factors such as user behavior characteristics and popularity of interest points are comprehensively considered in the analysis unit, and a similarity matrix between user-interest points is constructed using advanced algorithms and models. Finally, a personalized recommendation list is generated in the list generation module based on multiple factors.
It improves the accuracy and personalization of the recommendation system, can more accurately meet users' diverse needs, and improves user experience and satisfaction.
Smart Images
Figure CN119988759A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of interest point recommendation, and in particular to a method and system for recommending interest points integrating multiple factors. Background Art
[0002] In the current information society, the point of interest recommendation system has become an indispensable part of people's daily life. It is widely used in many fields such as tourism, catering, and entertainment.
[0003] However, most existing POI recommendation systems make recommendations based only on single factors such as the user's geographic location and historical behavior, ignoring the diversity and complexity of user POIs as well as multi-dimensional information such as time and space. As a result, the recommendation results are often not accurate enough and cannot meet the user's personalized needs.
[0004] To this end, those skilled in the art have proposed a method and system for recommending points of interest that integrate multiple factors to solve the problems raised by the background technology. Summary of the invention
[0005] In order to solve the above technical problems, the present invention provides a method and system for recommending points of interest that integrate multiple factors, so as to solve the problem that most existing points of interest recommendation systems in the prior art only make recommendations based on single factors such as the user's geographical location and historical behavior, ignoring the diversity and complexity of the user's points of interest as well as multi-dimensional information such as time and space, resulting in the recommendation results being often not accurate enough and unable to meet the user's personalized needs.
[0006] A method for recommending points of interest by integrating multiple factors, comprising:
[0007] S1. The user name collection module in the information collection unit identifies the user ID, the location collection module collects the location information of the check-in and clock-in under the user ID, the image scanning module scans and identifies the image information of the clock-in, and the location classification module classifies and identifies the location of the clock-in according to the image scan information, and then transmits the collected data to the analysis unit;
[0008] S2, the rating collection module collects the rating data of the user's check-in location, the time collection module collects the time of the user's check-in, the popularity collection module collects the popularity data of the check-in location, the spatial distance collection module and the like record collection module collect data on the spatial distance of the point of interest and the user's like record, and transmits the collected data factors to the central processing system;
[0009] S3, the central processing system then transmits the data to the modeling unit, the receiving module receives the transmitted data, the model generation module establishes the model according to the uploaded influencing factor data, constructs the similarity matrix, the calculation module obtains the factors that can affect the user, and then uploads the calculated result data to the central processing system;
[0010] S4, the information collection unit uploads the continuously collected check-in location data to the database of the data processing unit for storage, the integration module statistically organizes the data in the database, and the classification module classifies the data in the database, specifically classifying points of interest with similar location properties into one category. When the check-in location no longer exists, the deletion module deletes the data in the database accordingly;
[0011] S5. The extraction module extracts points of interest with similar properties from the database according to the user's favorite factors, and the list generation module lists the extracted location data in a table to display to the user for selection.
[0012] Preferably, the information collection unit includes a user name collection module, a location collection module, a picture scanning module and a location classification module.
[0013] Preferably, the analysis unit includes a score collection module, a time collection module, a popularity collection module, a spatial distance collection module and a like record collection module.
[0014] Preferably, in the rating collection module, a weighted average method is used to integrate the ratings of multiple users on the points of interest.
[0015] Preferably, the modeling unit includes a receiving module, a model generating module and a calculating module; the calculating module uses a cosine similarity algorithm when calculating the similarity between the user and the point of interest;
[0016] In the modeling unit, in order to more accurately construct the similarity matrix between users and interest points, the cosine similarity optimization algorithm is introduced to help the system more accurately calculate the similarity between users or the similarity between interest points, thereby improving the accuracy of recommendations.
[0017] Preferably, the data processing unit includes a database, the input end of the database is electrically connected to the output end of the deletion module, the input end of the database is electrically connected to the output end of the integration module, and the input end of the database is electrically connected to the output end of the classification module.
[0018] Preferably, in step S4, the integration module statistically organizes the data in the database, the classification module classifies the data in the database, and introduces a clustering algorithm to classify the points of interest.
[0019] Preferably, in step S5, in order to mine the potential relationship between the user and the interest point, a variant algorithm of singular value decomposition (SVD) is introduced.
[0020] Preferably, the list generating module combines multiple factors for weighted summation when generating the recommendation list.
[0021] A point of interest recommendation system integrating multiple factors includes a central processing system, and uses the above-mentioned point of interest recommendation method integrating multiple factors, including: the central processing system is bidirectionally connected to an information collection unit via wireless, the output end of the information collection unit is electrically connected to the input end of an analysis unit, the output end of the analysis unit is electrically connected to the input end of the central processing system, the central processing system is bidirectionally connected to a data processing unit via wireless, the central processing system is bidirectionally connected to a modeling unit via wireless, the output end of the central processing system is electrically connected to the input end of an extraction module, and the output end of the extraction module is electrically connected to the input end of a list generation module.
[0022] A processor is configured to execute the above-mentioned method for recommending points of interest by integrating multiple factors.
[0023] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the above-mentioned method for recommending points of interest by integrating multiple factors.
[0024] Compared with the prior art, the present invention has the following beneficial effects:
[0025] 1. The present invention comprehensively and accurately collects the user's point of interest information, including user name, location, pictures and other data, through the information collection unit, providing a solid foundation for subsequent analysis and recommendation; this solves the problem of incomplete and inaccurate data collection in the prior art and improves the reliability and accuracy of the recommendation system.
[0026] 2. The present invention introduces a variety of data collection modules in the analysis unit, such as a rating collection module, a time collection module, a popularity collection module, etc., which can comprehensively consider multiple factors such as user behavior characteristics and the popularity of points of interest, thereby constructing a more accurate user portrait and point of interest features; this helps to improve the personalization and accuracy of the recommendation system.
[0027] 3. The present invention adopts advanced algorithms and models in the modeling unit, such as cosine similarity algorithm, clustering algorithm, etc., which can construct a model that can fully reflect the relationship between user interests and interest point features; this further improves the accuracy and personalization of the recommendation system, and provides users with recommendation results that are more in line with their needs.
[0028] 4. The present invention also integrates, classifies and deletes the data in the database through the data processing unit to ensure the accuracy and timeliness of the data; at the same time, it introduces advanced technologies such as variant algorithms of singular value decomposition to further explore the potential relationship between users and points of interest, thereby improving the accuracy of recommendations and user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] Figure 1 This is a flow chart of the method for recommending points of interest integrating multiple factors according to the present invention. DETAILED DESCRIPTION
[0030] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.
[0031] Embodiment: The present invention provides a method for recommending points of interest by integrating multiple factors, such as Figure 1 As shown, including:
[0032] S1. The user name collection module in the information collection unit identifies the user ID, the location collection module collects the location information of the check-in and clock-in under the user ID, the image scanning module scans and identifies the image information of the clock-in, and the location classification module classifies and identifies the location of the clock-in according to the image scan information, and then transmits the collected data to the analysis unit;
[0033] S2, the rating collection module collects the rating data of the user's check-in location, the time collection module collects the time of the user's check-in, the popularity collection module collects the popularity data of the check-in location, the spatial distance collection module and the like record collection module collect data on the spatial distance of the point of interest and the user's like record, and transmits the collected data factors to the central processing system;
[0034] S3, the central processing system then transmits the data to the modeling unit, the receiving module receives the transmitted data, the model generation module establishes the model according to the uploaded influencing factor data, constructs the similarity matrix, the calculation module obtains the factors that can affect the user, and then uploads the calculated result data to the central processing system;
[0035] S4, the information collection unit uploads the continuously collected check-in location data to the database of the data processing unit for storage, the integration module statistically organizes the data in the database, and the classification module classifies the data in the database, specifically classifying points of interest with similar location properties into one category. When the check-in location no longer exists, the deletion module deletes the data in the database accordingly;
[0036] S5. The extraction module extracts points of interest with similar properties from the database according to the user's favorite factors, and the list generation module lists the extracted location data in a table to display to the user for selection.
[0037] From the above, it can be seen that this method can construct a more accurate user portrait and interest point features by comprehensively and accurately collecting various interest point information of users, and comprehensively considering multiple factors such as ratings, time, popularity, spatial distance and like records, thereby providing users with more personalized and accurate recommendation results; at the same time, through the application of advanced algorithms and models, as well as data integration, classification and potential relationship mining, the present invention further improves the accuracy of recommendations and user satisfaction.
[0038] Furthermore, the information collection unit includes a user name collection module, a location collection module, a picture scanning module and a location classification module.
[0039] From the above, we can see that the system can comprehensively and meticulously collect user's POI information, including user identity, specific location, check-in pictures and place classification, etc., providing a rich and accurate data basis for subsequent analysis and recommendation, which helps to improve the reliability and accuracy of the recommendation system and bring users a more accurate and personalized recommendation experience.
[0040] Furthermore, the analysis unit includes a score collection module, a time collection module, a popularity collection module, a spatial distance collection module and a like record collection module.
[0041] From the above, we can see that it is possible to capture users' multi-dimensional feedback on points of interest in an all-round way, including ratings, visit time, popularity, spatial distance considerations, and social interactions (such as likes), thereby constructing richer and more three-dimensional user point of interest features and behavior patterns, providing in-depth and accurate data support for subsequent modeling and recommendations, helping to improve the personalization and accuracy of recommendations, and better meet the diverse needs of users.
[0042] Furthermore, in the rating collection module, a weighted average method is used to integrate the ratings of multiple users on the points of interest. The weighted average method includes: assuming R ui is the rating of user u on point of interest i, n is the number of ratings, then the weighted average rating is:
[0043]
[0044] Among them, w u is the weight of user u.
[0045] From the above, we can see that in the rating collection module, the weighted average method is used to integrate the ratings of multiple users on the points of interest and take into account the user's weight, which can more reasonably reflect the influence of different users on the evaluation of the points of interest and avoid the deviation of the overall evaluation caused by a single user or extreme rating; by assigning appropriate weights to different users, the weighted average method can more accurately calculate the comprehensive score of the point of interest, thereby helping the recommendation system to more accurately evaluate the quality and popularity of the point of interest and provide users with a more reliable recommendation basis.
[0046] Furthermore, the modeling unit includes a receiving module, a model generating module and a calculating module; the calculating module uses a cosine similarity algorithm when calculating the similarity between the user and the point of interest, and the formula of the cosine similarity algorithm is as follows:
[0047]
[0048] in, and are the feature vectors of users and points of interest respectively;
[0049] In the modeling unit, in order to more accurately construct the similarity matrix between users and interest points, the cosine similarity optimization algorithm is introduced to help the system more accurately calculate the similarity between users or the similarity between interest points, thereby improving the accuracy of recommendation. The formula of the cosine similarity optimization algorithm is as follows:
[0050]
[0051] Among them, u i and p i Represents the i-th feature value of the user and the point of interest respectively, w i is the corresponding weight factor.
[0052] From the above, it can be seen that in the modeling unit, a combination of a receiving module, a model generation module and a calculation module is used, and the cosine similarity algorithm and its optimization algorithm are introduced to calculate the similarity between users and points of interest. The cosine similarity algorithm can effectively measure the similarity and correlation between them based on the feature vectors of users and points of interest; and by introducing the cosine similarity optimization algorithm, the weight factor of the feature value is further considered, making the calculation of the similarity more accurate and practical, which not only improves the recommendation system's ability to understand user preferences and points of interest characteristics, but also can more accurately explore the potential connection between users and points of interest, thereby providing users with more accurate recommendation results that meet their personalized needs, significantly improving the accuracy of recommendations and user satisfaction.
[0053] Furthermore, the data processing unit includes a database, an input end of the database is electrically connected to an output end of the deletion module, an input end of the database is electrically connected to an output end of the integration module, and an input end of the database is electrically connected to an output end of the classification module.
[0054] From the above, we can see that the real-time, accuracy and completeness of the data in the database are ensured; through the integration module, the database can receive and integrate data from different sources to form a unified data view; the classification module further classifies the data scientifically to facilitate subsequent data management and analysis; at the same time, the deletion module can clean up outdated or invalid data in a timely manner to prevent data redundancy and the accumulation of erroneous information; this design not only improves the efficiency of data processing, but also provides more reliable and valuable data support for the recommendation system, thereby improving the accuracy of recommendations and user experience.
[0055] Further, in step S4, the integration module statistically organizes the data in the database, the classification module classifies the data in the database, and introduces a clustering algorithm to classify the points of interest, including: X is the feature matrix of the interest points, k is the number of clusters, then the objective function of the clustering algorithm is:
[0056]
[0057] Where C={C1,C2,...,C k} is the clustering partition, μ i is cluster C i The center of mass.
[0058] From the above, we can see that by setting the feature matrix and number of clusters of points of interest, the clustering algorithm can automatically discover the inherent structure and patterns in the data and classify points of interest with similar characteristics or attributes into one category; this not only simplifies the complexity of the data set, but also helps the recommendation system to have a deeper understanding of user preferences and the characteristics of points of interest; based on the clustering results, the system can more effectively explore the potential connections between users and points of interest, and provide more personalized and accurate recommendation services, thereby improving user experience and satisfaction.
[0059] Furthermore, in step S5, in order to mine the potential relationship between the user and the interest point, a variant algorithm of singular value decomposition (SVD) is introduced, and the variant algorithm of singular value decomposition is as follows:
[0060]
[0061] Where R is the user-interest point matrix, U and V are the latent factor matrices of users and interest points respectively, Σ is a diagonal matrix, and λ U and λ V is the regularization parameter.
[0062] From the above, we can see that this variant algorithm can reveal the latent factors hidden behind users and points of interest by decomposing the user-point of interest matrix. These latent factors reflect the essential characteristics of user preferences and the unique attributes of points of interest. The introduction of regularization parameters enhances the stability and generalization ability of the algorithm and avoids the overfitting problem. Based on these latent factors, the recommendation system can more accurately predict the user's preference for unknown points of interest and discover the potential connection between users and points of interest, thereby providing users with more personalized recommendation results that meet their potential needs, significantly improving the accuracy of recommendations and user satisfaction.
[0063] Furthermore, the list generation module, when generating the recommendation list, combines multiple factors for weighted summation, including: assuming s geo 、s so 、s tem_geo are the geographical influence score, social influence score and spatiotemporal influence score respectively, λ1, λ2 and λ3 are the corresponding weight coefficients, and the comprehensive preference score is:
[0064] s uv =λ1s geo +λ2s so +λ3s tem_geo ;
[0065] Among them, s uv is the comprehensive preference score of user u for point of interest v.
[0066] From the above, we can see that by comprehensively considering multiple influencing factors and assigning them corresponding weight coefficients, this module can more comprehensively evaluate the user's preference for points of interest; this approach not only improves the personalization and accuracy of recommendations, but also better meets the diverse needs of users in different scenarios; for example, for users who pay attention to geographical location, the geographical influence score will occupy a larger weight, thereby ensuring that the recommended points of interest are close to the user's current location or preferred location; for users who are more influenced by their social circles, the social influence score will occupy an important position in the comprehensive preference score; this comprehensive evaluation method makes the recommendation results more in line with the user's actual needs and preferences, and improves the user's usage experience and satisfaction.
[0067] A point of interest recommendation system integrating multiple factors includes a central processing system, and uses the above-mentioned point of interest recommendation method integrating multiple factors, including: the central processing system is bidirectionally connected to an information collection unit via wireless, the output end of the information collection unit is electrically connected to the input end of an analysis unit, the output end of the analysis unit is electrically connected to the input end of the central processing system, the central processing system is bidirectionally connected to a data processing unit via wireless, the central processing system is bidirectionally connected to a modeling unit via wireless, the output end of the central processing system is electrically connected to the input end of an extraction module, and the output end of the extraction module is electrically connected to the input end of a list generation module.
[0068] Working principle: This method comprehensively collects user interest point information through the information collection unit. The analysis unit comprehensively considers various factors to build user portraits and interest point features. The modeling unit uses advanced algorithms and models to calculate the similarity between users and interest points. The data processing unit integrates, classifies and deletes the data in the database to ensure data accuracy. Finally, the list generation module combines various factors to generate a personalized recommendation list. The entire system realizes data transmission and coordination between units through the central processing system, thereby providing users with accurate and personalized interest point recommendation services.
[0069] Furthermore, the effect of a method for recommending points of interest integrating multiple factors in the embodiment is compared with the existing point of interest recommendation system (comparative example), and the following table is obtained:
[0070]
[0071]
[0072] It can be seen from the above table that the fusion multi-factor POI recommendation method of the embodiment is superior to the existing POI recommendation system in terms of data collection, user portrait construction, algorithms and models, data processing, recommendation accuracy and user experience; therefore, the method can provide users with more personalized and accurate recommendation services, and improve user experience and satisfaction.
[0073] The above units and modules and their uses are as follows:
[0074] Information collection unit
[0075] Purpose: Responsible for comprehensively collecting users' points of interest information, including user ID, location information, image information, etc., to provide a data basis for subsequent analysis and recommendations.
[0076] Analysis Unit
[0077] Purpose: To conduct multi-dimensional analysis on the collected user POI information, including ratings, time, popularity, spatial distance, and likes records, in order to construct user portraits and POI features.
[0078] Modeling Unit
[0079] Purpose: Use advanced algorithms and models (such as cosine similarity algorithm, clustering algorithm, etc.) to calculate the similarity between users and points of interest, build a user-point of interest similarity matrix, and provide a basis for recommendation.
[0080] Data processing unit
[0081] Purpose: To integrate, classify, and delete data in the database to ensure the accuracy and timeliness of the data and provide reliable data support for the recommendation system.
[0082] Recommendation list generation unit
[0083] Purpose: Based on the user's favorite factors, extract points of interest with similar properties from the database, combine multiple factors for weighted summation, and generate a personalized recommendation list to display to the user
[0084] Username collection module
[0085] Purpose: To identify the user ID and ensure that the collected POI information corresponds to the user identity.
[0086] Position acquisition module
[0087] Purpose: Collect the location information of check-in and clock-in under the user ID to provide location basis for location classification and recommendation.
[0088] Image scanning module
[0089] Purpose: Scan and identify the image information of the check-in to obtain more detailed information about the point of interest.
[0090] Location Classification Module
[0091] Purpose: To classify and identify check-in locations based on the information from image scans to improve the accuracy of location recognition.
[0092] Rating collection module
[0093] Purpose: Collect user ratings of check-in locations to assess the popularity of points of interest.
[0094] Time acquisition module
[0095] Purpose: Collect user check-in time information to analyze the time patterns of user behavior and the time characteristics of points of interest.
[0096] Popularity collection module
[0097] Purpose: Collect popularity data of check-in locations to reflect the popularity and popularity of points of interest.
[0098] Spatial distance acquisition module
[0099] Purpose: Collect spatial distance information of points of interest to consider the impact of geographic location on recommendation results.
[0100] Like record collection module
[0101] Purpose: Collect users' likes records for analyzing their social interaction behaviors and interest preferences.
[0102] Receiver Module
[0103] Purpose: To receive data transmitted from the analysis unit in the modeling unit.
[0104] Model generation module
[0105] Purpose: Generate a recommendation model based on the uploaded influencing factor data and construct a similarity matrix.
[0106] Computing Module
[0107] Purpose: Calculate the similarity between users and points of interest to determine the factors that can influence users.
[0108] Integration Module
[0109] Purpose: To organize statistics of data in the database and form a unified data view.
[0110] Classification Module
[0111] Purpose: To classify the data in the database and group points of interest with similar location properties into one category.
[0112] Deleting a module
[0113] Purpose: When the check-in location no longer exists, the corresponding data in the database will be deleted to ensure the timeliness of the data.
[0114] Extraction module
[0115] Purpose: Extract points of interest with similar properties from the database based on the user's favorite factors.
[0116] List generation module
[0117] Purpose: To present the extracted location data in a table to the user for selection.
[0118] The present application embodiment provides an electronic device applicable to the above-mentioned method for recommending points of interest integrating multiple factors, including:
[0119] Memory, used to protect computer programs and data;
[0120] Processor, used to run system programs.
[0121] An embodiment of the present application provides a computer storage medium, which is applicable to the above-mentioned method for recommending points of interest integrating multiple factors, and performs hierarchical confidentiality management on the above-mentioned system and data in accordance with confidentiality management requirements.
[0122] Those skilled in the art will appreciate that the embodiments of the present application may be provided as a system or a computer program product. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.
[0123] The present application is described with reference to the flowcharts and / or block diagrams of the devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0124] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0125] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0126] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0127] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0128] Computer readable media include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0129] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, commodity or device including the elements.
[0130] The embodiments of the present invention are provided for the purpose of illustration and description. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations of the present invention. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present invention.
Claims
1. A method for recommending points of interest by integrating multiple factors, characterized in that: include: S1. The user name collection module in the information collection unit identifies the user ID, the location collection module collects the location information of the check-in and clock-in under the user ID, the image scanning module scans and identifies the image information of the clock-in, and the location classification module classifies and identifies the location of the clock-in according to the image scan information, and then transmits the collected data to the analysis unit; S2, the rating collection module collects the rating data of the user's check-in location, the time collection module collects the time of the user's check-in, the popularity collection module collects the popularity data of the check-in location, the spatial distance collection module and the like record collection module collect data on the spatial distance of the point of interest and the user's like record, and transmits the collected data factors to the central processing system; S3, the central processing system then transmits the data to the modeling unit, the receiving module receives the transmitted data, the model generation module establishes the model according to the uploaded influencing factor data, constructs the similarity matrix, the calculation module obtains the factors that can affect the user, and then uploads the calculated result data to the central processing system; S4, the information collection unit uploads the continuously collected check-in location data to the database of the data processing unit for storage, the integration module statistically organizes the data in the database, and the classification module classifies the data in the database, specifically classifying points of interest with similar location properties into one category. When the check-in location no longer exists, the deletion module deletes the data in the database accordingly; S5. The extraction module extracts points of interest with similar properties from the database according to the user's favorite factors, and the list generation module lists the extracted location data in a table to display to the user for selection.
2. A method for recommending points of interest by integrating multiple factors as claimed in claim 1, characterized in that: The information collection unit includes a user name collection module, a location collection module, a picture scanning module and a location classification module.
3. A method and system for recommending points of interest integrating multiple factors as claimed in claim 1, characterized in that: The analysis unit includes a score collection module, a time collection module, a popularity collection module, a spatial distance collection module and a like record collection module.
4. The method for recommending points of interest by integrating multiple factors as claimed in claim 1, characterized in that: In the rating collection module, the weighted average method is used to integrate the ratings of multiple users on the points of interest.
5. The method for recommending points of interest by integrating multiple factors as claimed in claim 1, characterized in that: The modeling unit includes a receiving module, a model generating module and a calculating module; the calculating module uses a cosine similarity algorithm when calculating the similarity between the user and the point of interest; In the modeling unit, in order to more accurately construct the similarity matrix between users and interest points, the cosine similarity optimization algorithm is introduced to help the system more accurately calculate the similarity between users or the similarity between interest points.
6. The method for recommending points of interest by integrating multiple factors as claimed in claim 1, characterized in that: The data processing unit comprises a database, an input end of the database is electrically connected to an output end of the deletion module, an input end of the database is electrically connected to an output end of the integration module, and an input end of the database is electrically connected to an output end of the classification module.
7. The method for recommending points of interest by integrating multiple factors as claimed in claim 1, characterized in that: In step S4, the integration module statistically organizes the data in the database, the classification module classifies the data in the database, and introduces a clustering algorithm to classify the points of interest.
8. The method for recommending points of interest by integrating multiple factors as claimed in claim 1, characterized in that: In step S5, in order to mine the potential relationship between users and points of interest, a variant algorithm of singular value decomposition is introduced.
9. The method for recommending points of interest by integrating multiple factors as claimed in claim 1, characterized in that: The list generation module combines multiple factors for weighted summation when generating the recommendation list.
10. A point of interest recommendation system integrating multiple factors, comprising a central processing system, characterized in that: The method for recommending points of interest by integrating multiple factors as described in any one of claims 1 to 9 comprises: the central processing system is bidirectionally connected to the information collection unit via wireless, the output end of the information collection unit is electrically connected to the input end of the analysis unit, the output end of the analysis unit is electrically connected to the input end of the central processing system, the central processing system is bidirectionally connected to the data processing unit via wireless, the central processing system is bidirectionally connected to the modeling unit via wireless, the output end of the central processing system is electrically connected to the input end of the extraction module, and the output end of the extraction module is electrically connected to the input end of the list generation module.