Big data-based Beidou navigation location service recommendation system and method

Abstract

The invention relates to the technical field of Beidou navigation, in particular to a Beidou navigation position service recommendation system and method based on big data, and the method comprises the following steps: S1, Beidou multi-source data collection: obtaining the Beidou multi-source data of a user through a Beidou satellite navigation system; s2, dynamic spatio-temporal feature coding: constructing spatio-temporal feature vectors; s3, privacy protection feature extraction: generating a standardized feature data set; s4, user portrait generation: generating a user portrait including a real-time demand weight; s5, multi-target recommendation generation: establishing a multi-target optimization model including a service matching degree, a privacy protection degree and a resource consumption coefficient, and generating a recommendation scheme in combination with the generated user portrait; according to the method, the integrity and availability of the data are improved, and the user behavior mode, the environment dynamic factor and the service adaptation degree are fully considered in the data level, so that the recommendation result better meets the actual demand of the user.

Description

Technical Field

[0001] The present invention relates to the field of Beidou navigation technology, and in particular to a Beidou navigation location service recommendation system and method based on big data. Background Art

[0002] With the widespread application of the Beidou satellite navigation system, location-based services have become an important supporting technology in the fields of intelligent transportation, logistics scheduling, smart travel and emergency rescue. Traditional location-based services mainly rely on GNSS data for user positioning and combine with geographic information systems to provide basic navigation and path planning functions. However, with the continuous development of big data, artificial intelligence and personalized service needs, users not only need accurate geographic location information, but also hope to obtain personalized recommendation services that meet their needs, such as intelligent travel suggestions, point of interest (POI) recommendations, and environmental adaptation navigation. Therefore, how to provide accurate, efficient and secure intelligent recommendation services based on the multi-source data of the Beidou navigation system and combined with the personalized behavior characteristics of users has become an important direction of current research.

[0003] Existing navigation location service recommendation methods mainly rely on static path planning or recommendation algorithms based on historical behavior, but there are many deficiencies in data fusion, personalized recommendation and privacy protection. First, the data utilization rate is low. Existing methods usually only rely on GNSS trajectory data and fail to fully integrate multi-source data such as surrounding POI, meteorological environment, and user historical behavior, resulting in low accuracy and lack of real-time recommendation results. Secondly, the personalized recommendation ability is insufficient. Most existing recommendation systems are based on rule matching or simple similarity calculations, and fail to effectively combine user portraits and real-time demand weights, resulting in a lack of targeted recommendations and inability to dynamically adapt to changes in user needs. In addition, there is a lack of privacy protection mechanism. Traditional location recommendation systems usually directly store and calculate user trajectory data, which poses a serious risk of privacy leakage. The lack of an effective privacy protection mechanism makes it difficult to ensure recommendation accuracy while taking into account user data security. Finally, the computing resource consumption is high. When optimizing recommendations, existing methods fail to fully consider system resource utilization, resulting in high computing overhead and difficulty in applying to large-scale application scenarios.

[0004] In view of the deficiencies in the prior art, the present invention proposes a Beidou navigation location service recommendation system and method based on big data, which improves the intelligence level and user experience of Beidou navigation location service, makes the recommendation results more accurate and reliable, and has strong real-time adaptability. Summary of the invention

[0005] The present invention provides a Beidou navigation location service recommendation system and method based on big data.

[0006] The Beidou navigation location service recommendation method based on big data includes the following steps: S1, Beidou multi-source data acquisition: Obtain users' Beidou multi-source data through the Beidou satellite navigation system, including real-time positioning data, user historical trajectory data, surrounding POI feature data and meteorological environment data; S2, dynamic spatiotemporal feature coding: perform spatiotemporal correlation analysis on Beidou multi-source data and construct spatiotemporal feature vectors including position precision factor, motion state vector, and environmental feature matrix; S3, privacy-preserving feature extraction: Apply the differential privacy algorithm to reduce the dimension of the spatiotemporal feature vector to generate a standardized feature dataset with privacy protection; S4, user portrait generation: Based on the standardized feature data set, dynamic label matching is performed in combination with user behavior logs to generate user portraits including real-time demand weights; S5, multi-objective recommendation generation: establish a multi-objective optimization model including service matching degree, privacy protection degree, and resource consumption coefficient, and generate a recommendation plan based on the generated user portrait.

[0007] Optionally, the Beidou multi-source data acquisition in S1 includes: S11, Beidou real-time positioning data acquisition: through the positioning service of the Beidou satellite navigation system, receive the GNSS signal of the user terminal device to obtain the latitude and longitude coordinates, elevation and speed information of the current time; S12, user historical trajectory data collection: based on the continuous positioning data of the user terminal, the trajectory is stored in combination with the timestamp, and the user's movement mode, stop points and historical travel trajectory are extracted; S13, obtaining surrounding POI feature data: using the user's current location and historical trajectory data, calling the geographic information database or map service interface to obtain surrounding point of interest (POI) information, including transportation hubs, commercial areas, medical facilities, and associating user visit frequency and preference characteristics; S14, meteorological environment data collection: obtain the weather conditions, temperature and humidity, wind speed, visibility and electromagnetic interference of the user's current location through a third-party meteorological data interface.

[0008] Optionally, the dynamic spatiotemporal feature encoding in S2 includes: S21, Position DPI calculation: Calculate the user's position DPI based on real-time positioning information from Beidou multi-source data ; S22, motion state vector construction: Based on the user's historical trajectory data, extract the user's speed, acceleration, and heading angle to construct a motion state vector ; S23, environmental feature matrix construction: Based on the meteorological environment data of the user's current location and the surrounding POI information, an environmental feature matrix is ​​constructed. ; S24, spatiotemporal feature vector construction: The position precision factor, motion state vector and environmental feature matrix are integrated to form a spatiotemporal feature vector .

[0009] Optionally, the privacy protection feature extraction in S3 includes: S31, spatiotemporal feature vector dimensionality reduction preprocessing: principal component analysis (PCA) is used to reduce the spatiotemporal feature vector Perform dimensionality reduction to obtain the space-time feature vector after dimensionality reduction ; S32, Laplace noise perturbation: spatiotemporal feature vectors after dimensionality reduction Introduce Laplace noise to obtain the noisy spatiotemporal feature vector ; S33, Privacy-preserving feature data standardization: The spatiotemporal feature vector after noise addition Perform normalization to obtain a standardized feature data set .

[0010] Optionally, the user portrait generation in S4 includes: S41, User feature vector construction: based on standardized feature dataset and user historical behavior logs, extracting user personalized features and constructing user feature vectors ; S42, Dynamic Tag Matching: Based on User Feature Vector , combined with real-time behavior data, dynamic tag matching, and calculation of tag relevance ; S43, real-time demand weight calculation: tag association based on dynamic matching , calculate the weight of each requirement label ; S44, user portrait update: combine all requirement tags and their weights to generate the final user portrait .

[0011] Optionally, the multi-objective recommendation generation in S5 includes: S51, multi-objective optimization model construction: construct a multi-objective optimization model including service matching degree, privacy protection degree and resource consumption coefficient to balance recommendation accuracy, privacy security and resource utilization; S52, personalized recommendation solution generation: Based on the real-time demand weight of the user portrait, the candidate recommendation items are prioritized to generate a personalized recommendation solution.

[0012] Optionally, the multi-objective optimization model construction in S51 includes: S511, service matching calculation: Calculate the service matching based on the feature similarity between the user profile and the candidate recommendation items. ; S512, privacy protection calculation: Based on the differential privacy mechanism, calculate the privacy protection degree of the recommended solution for user data ; S513, resource consumption coefficient calculation: comprehensively consider computing cost, bandwidth usage and storage requirements, and calculate the resource consumption coefficient of the recommended solution ; S514, Multi-objective optimization model construction: combined with service matching , privacy protection and resource consumption coefficient , construct the multi-objective optimization objective function.

[0013] Optionally, the generation of the personalized recommendation scheme in S52 includes: S521, candidate recommendation item score calculation: Based on the real-time demand weight of the user portrait, the candidate recommendation items are scored and the comprehensive score of the candidate recommendation items is defined. ; S522, Prioritization and Personalized Recommendation: Based on Comprehensive Scoring Sort the candidate recommendation items and calculate the recommendation probability by combining Softmax normalization ; S523, Personalized recommendation scheme generation: based on the calculated recommendation probability , using the Top-K sampling strategy to generate the final personalized recommendation solution .

[0014] The Beidou navigation location service recommendation system based on big data is used to implement the above-mentioned Beidou navigation location service recommendation method based on big data, and includes the following modules: Beidou multi-source data acquisition module: obtains users' Beidou multi-source data through the Beidou satellite navigation system, including real-time positioning data, user historical trajectory data, surrounding POI feature data and meteorological environment data; Dynamic spatiotemporal feature coding module: performs spatiotemporal correlation analysis on Beidou multi-source data and constructs spatiotemporal feature vectors including position precision factor, motion state vector, and environmental feature matrix; Privacy-preserving feature extraction module: It applies differential privacy algorithm to reduce the dimension of spatiotemporal feature vectors and generates a standardized feature data set with privacy protection. User portrait generation module: Based on the standardized feature data set, dynamic label matching is performed in combination with user behavior logs to generate user portraits including real-time demand weights; Multi-objective recommendation generation module: establishes a multi-objective optimization model including service matching degree, privacy protection degree, and resource consumption coefficient, and generates a recommendation plan based on user portraits.

[0015] Beneficial effects of the present invention: The present invention constructs a Beidou navigation location service recommendation method with high precision, personalization, privacy protection and resource optimization through Beidou multi-source data collection, dynamic spatiotemporal feature coding, privacy protection feature extraction, user portrait generation and multi-target recommendation generation. The method comprehensively collects real-time positioning, historical trajectory, surrounding POI features and meteorological environment data, and generates accurate user location feature vectors through spatiotemporal correlation analysis and feature extraction, thereby improving the integrity and availability of data, and fully considering user behavior patterns, environmental dynamic factors and service adaptability at the data level, so that the recommendation results are more in line with the actual needs of users.

[0016] The present invention introduces a differential privacy mechanism and adds Laplace noise perturbation in the process of data dimensionality reduction to effectively protect user privacy. At the same time, it uses principal component analysis to reduce data dimensions and improve computational efficiency. In addition, based on user behavior logs and dynamic label matching, a real-time updated user portrait model is constructed to ensure that the recommendation system can quickly respond to changes in user needs and optimize the personalized recommendation effect.

[0017] The present invention adopts a multi-objective optimization model, takes into account service matching, privacy protection and resource consumption, ensures the balance between accuracy, security and computational efficiency of recommendation schemes, and optimizes and solves through Pareto optimal solution, so that the recommendation scheme can not only meet the personalized needs of users, but also reduce system resource consumption and ensure privacy security. In addition, combined with Softmax normalization and Top-K sampling strategy, the recommendation system can not only highlight high-matching options, but also take into account long-tail recommendations, improve the diversity and adaptability of recommendation results, thereby improving the intelligence level and user experience of Beidou navigation location services. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A schematic diagram of a recommended method flow chart of an embodiment of the present invention; Figure 2 Schematic diagram of system function modules according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0021] It should be noted that the references to "one embodiment", "embodiment", "exemplary embodiments", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0022] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0023] like Figure 1 As shown, the Beidou navigation location service recommendation method based on big data includes the following steps: S1, Beidou multi-source data acquisition: Obtain users' Beidou multi-source data through the Beidou satellite navigation system, including real-time positioning data, user historical trajectory data, surrounding POI feature data and meteorological environment data; S2, dynamic spatiotemporal feature coding: perform spatiotemporal correlation analysis on Beidou multi-source data and construct spatiotemporal feature vectors including position precision factor, motion state vector, and environmental feature matrix; S3, privacy-preserving feature extraction: Apply the differential privacy algorithm to reduce the dimension of the spatiotemporal feature vector to generate a standardized feature dataset with privacy protection; S4, user portrait generation: Based on the standardized feature data set, dynamic label matching is performed in combination with user behavior logs to generate user portraits including real-time demand weights; S5, multi-objective recommendation generation: establish a multi-objective optimization model including service matching degree, privacy protection degree, and resource consumption coefficient, and generate a recommendation plan based on the generated user portrait; Through the above content, high-precision, personalized, privacy-protected and environment-adaptive navigation service recommendations are achieved. It can comprehensively consider user behavior characteristics, service matching, and resource consumption, making the recommendation plan more accurate and reliable, and have strong real-time adaptability, thereby improving the intelligence level of navigation services and user experience.

[0024] Beidou multi-source data collection in S1 includes: S11, Beidou real-time positioning data acquisition: through the positioning service of the Beidou satellite navigation system, receive the GNSS signal of the user terminal device to obtain the latitude and longitude coordinates, elevation and speed information of the current time; S12, user historical trajectory data collection: based on the continuous positioning data of the user terminal, the trajectory is stored in combination with the timestamp, and the user's movement mode, stop points and historical travel trajectory are extracted; S13, obtaining surrounding POI feature data: using the user's current location and historical trajectory data, calling the geographic information database or map service interface to obtain surrounding point of interest (POI) information, including transportation hubs, commercial areas, medical facilities, and associating user visit frequency and preference characteristics; S14, meteorological environment data collection: obtain the weather conditions, temperature and humidity, wind speed, visibility and electromagnetic interference of the user's current location through the third-party meteorological data interface; Through the above content, real-time positioning, historical trajectory, surrounding POI characteristics and meteorological environment data are integrated to construct a high-precision, dynamically adaptable user location dataset, which not only improves the positioning accuracy and trajectory analysis accuracy, but also mines user interest preferences through POI data, and combines meteorological environment information to enhance the reliability and adaptability of services, thereby providing more accurate data support for personalized recommendations, optimizing user experience, and improving the intelligent application capabilities of the Beidou navigation system in complex environments.

[0025] The dynamic spatiotemporal feature encoding in S2 includes: S21, Position DPI calculation: Calculate the user's position DPI based on real-time positioning information from Beidou multi-source data , expressed as: ; in, is the position precision factor, is the number of anchor points, For the The coordinates of the positioning points, is the average coordinate of all positioning points; S22, motion state vector construction: Based on the user's historical trajectory data, extract the user's speed, acceleration, and heading angle to construct a motion state vector , expressed as: ; in, Indicates speed, is the distance between adjacent positioning points of the user, is the time interval, represents acceleration, is the speed change, represents the heading angle, and The coordinates of the current and next positioning points respectively; S23, environmental feature matrix construction: Based on the meteorological environment data of the user's current location and the surrounding POI information, an environmental feature matrix is ​​constructed. , expressed as: in, is the temperature, For humidity, For visibility, is the electromagnetic interference intensity, , , They are the 1st, 2nd, The characteristic values ​​of each point of interest (such as distance, type weight, etc.); Feature values ​​of interest points The calculation includes: (1) POI relevance score calculation: Calculate the geographical distance between the user's current location and the POI, and combine the user's historical visit frequency and the POI category weight to obtain the POI relevance score. , expressed as: ; ; in, The user's current location and The Euclidean distance between POIs, is the user's current location coordinates, for The coordinates of is the historical frequency of users visiting the POI, indicating the user's preference. is the POI category weight, indicating the importance of the POI in this category (such as transportation hubs, shopping malls, restaurants, etc.). , , are the influencing factors of distance, historical access frequency, and category weight, satisfying ; (2) POI normalization: In order to avoid the influence of the magnitude difference of different features on the calculation results, all POI relevance scores are normalized, expressed as: in, and are the minimum and maximum values ​​of all POI relevance scores, respectively; S24, spatiotemporal feature vector construction: The position precision factor, motion state vector and environmental feature matrix are integrated to form a spatiotemporal feature vector , expressed as: ; Through the above content, not only the positioning accuracy and trajectory analysis accuracy are improved, but also the user interest preferences are explored through POI correlation calculation, and the service matching degree is optimized in combination with environmental factors, thereby improving the personalization and intelligence level of navigation services. At the same time, comprehensive data normalization processing ensures the uniformity and comparability of different data sources, making the recommendation system more accurate and reliable, and enhancing the adaptability and user experience of the Beidou navigation system in complex scenarios.

[0026] Privacy-preserving feature extraction in S3 includes: S31, spatiotemporal feature vector dimensionality reduction preprocessing: principal component analysis (PCA) is used to reduce the spatiotemporal feature vector Perform dimensionality reduction to reduce computational complexity and remove redundant information to obtain the reduced-dimensional spatiotemporal feature vector , expressed as: ; in, is the space-time feature vector, It is the dimension reduction transformation matrix calculated by PCA; S32, Laplace noise perturbation: To protect user privacy, the spatiotemporal feature vector after dimensionality reduction is Introduce Laplace noise to obtain the noisy spatiotemporal feature vector , so that the data satisfies the differential privacy constraint when querying, expressed as: ; ; ; in, is the spatiotemporal feature vector after noise addition, represents the noise that follows the Laplace distribution, is the scale parameter, which determines the shape and diffusion of the Laplace distribution. is the global sensitivity of the feature data, that is, the maximum possible range of variation, To balance privacy protection and data availability for privacy budget, is the probability density function of the Laplace distribution, is a random variable, i.e., the generated Laplace noise value; S33, privacy-preserving feature data standardization: To ensure that data of different feature dimensions have the same scale, the noisy spatiotemporal feature vectors are normalized. Perform normalization to obtain a standardized feature data set , expressed as: in, and are the minimum and maximum values ​​in the noise-added feature vector, respectively; Through the above content, on the basis of dimensionality reduction of spatiotemporal feature data, Laplace noise perturbation is introduced to effectively prevent the leakage of user sensitive information while maintaining data availability. The global sensitivity is used to control the noise intensity to achieve a balance between privacy protection and data accuracy. The stability and computational efficiency of the model are improved through standardized processing, which can provide stronger anti-attack capabilities in complex spatiotemporal data environments, so that Beidou navigation location service recommendations can not only meet personalized needs but also ensure user privacy and security.

[0027] User portrait generation in S4 includes: S41, User feature vector construction: based on standardized feature dataset and user historical behavior logs, extracting user personalized features and constructing user feature vectors , expressed as: ; ; in, To standardize the feature dataset, is the user behavior feature vector, which represents the user’s past interaction behavior. is the frequency vector of POI types visited by users, is the user's preference in different time periods, The user's historical interaction weight (such as clicks, stay time, number of visits, etc.); S42, Dynamic Tag Matching: Based on User Feature Vector , combined with real-time behavior data, dynamic tag matching, and calculation of tag relevance , expressed as: ; in, For the current moment The label relevance of , Cosine similarity, is the feature vector of the candidate label, , are the weight coefficients of spatiotemporal features and user behavior features, ; The calculation formula for cosine similarity is: ; in, is the vector dot product, , is the vector modulus; S43, real-time demand weight calculation: tag association based on dynamic matching , calculate the weight of each requirement label , expressed as: in, Requirement tag The real-time weight of is the current matching degree, is the temperature parameter, which controls the steepness of the weight distribution. is the matching degree between the user feature vector and the candidate label vector, is a mathematical constant; S44, user portrait update: combine all requirement tags and their weights to generate the final user portrait , expressed as: in, Create user portraits to represent the user's current personalized needs and preferences, and serve as input for the recommendation algorithm to optimize the accuracy and personalization of the recommendation service; Through the above content, it is possible to accurately capture the user's interest preferences and make adaptive adjustments based on real-time interactive behaviors to improve the accuracy and response speed of the recommendation system. At the same time, the cosine similarity is used to calculate the matching degree, and the weight distribution is adjusted by the temperature parameter, so that the recommendation can not only highlight the highly relevant services, but also appropriately retain the long-tail demand, enhance the personalization, stability and adaptability of the recommendation, thereby optimizing the user experience and improving the intelligence level of Beidou navigation location services.

[0028] The multi-objective recommendation generation in S5 includes: S51, multi-objective optimization model construction: construct a multi-objective optimization model including service matching degree, privacy protection degree and resource consumption coefficient to balance recommendation accuracy, privacy security and resource utilization; S52, personalized recommendation solution generation: based on the real-time demand weight of the user portrait, the candidate recommendation items are prioritized and a personalized recommendation solution is generated; Through the above content, a balance is achieved in the accuracy, security and efficiency of the recommendation scheme. Combined with the real-time demand weight of the user portrait, the candidate recommendation items are prioritized to ensure that the recommendation results are more personalized and dynamically adapt to user needs. This not only improves the accuracy of the recommendation and user experience, but also effectively reduces the consumption of computing resources and enhances privacy protection capabilities, making the Beidou navigation and location service more intelligent, adaptive and secure in complex environments.

[0029] The multi-objective optimization model construction in S51 includes: S511, service matching calculation: Calculate the service matching based on the feature similarity between the user profile and the candidate recommendation items. , expressed as: in, is the feature weight, Candidate recommendation The characteristic vector of is the cosine similarity, is the feature dimension parameter, i.e., the number of features of user portraits and candidate recommendation items; S512, privacy protection calculation: Based on the differential privacy mechanism, calculate the privacy protection degree of the recommended solution for user data , expressed as: ; in, is the privacy protection degree, the value range is , Budget for privacy; S513, resource consumption coefficient calculation: comprehensively consider computing cost, bandwidth usage and storage requirements, and calculate the resource consumption coefficient of the recommended solution , expressed as: ; in, is the resource consumption coefficient, is the computational cost (CPU / GPU computation time), is the bandwidth consumption (data transfer size), For storage requirements (recommended cache size), , , is the weight coefficient; S514, Multi-objective optimization model construction: combined with service matching , privacy protection and resource consumption coefficient , construct the multi-objective optimization objective function, expressed as: ; in, , , is the weight coefficient, To optimize the objective function, the Pareto optimal solution is used to ensure that the recommendation solution achieves the best balance between accuracy, privacy protection and resource utilization; The Pareto optimal solution is solved by the weighted sum method, which is expressed as: ; in, To optimize the target value, For the The objective function value, is the target weight, is the ideal optimal value for each objective, expressed as ; The objective function value specifically includes: Service matching target: (maximize the matching degree, take the negative sign when solving the minimization problem); Privacy protection goals: (maximize privacy protection, take the negative sign); Resource consumption minimization goal: (Minimize resource consumption and use the original value directly); Through the above content, we ensure that the recommendation system achieves the optimal balance between accuracy, security and computing cost. By adopting the Pareto optimal solution method, the recommendation scheme can not only meet personalized needs, but also effectively reduce resource consumption, while ensuring user privacy and security.

[0030] The generation of personalized recommendation solutions in S52 includes: S521, candidate recommendation item score calculation: Based on the real-time demand weight of the user portrait, the candidate recommendation items are scored and the comprehensive score of the candidate recommendation items is defined. , expressed as: in, The real-time demand weight of the user portrait, is the comprehensive score of the candidate recommendation items, For candidate recommendations in the Requirements tab The service matching degree on For candidate recommendations in the Requirements tab The privacy protection level, For candidate recommendations in the Requirements tab The resource consumption coefficient on , , are the corresponding weight parameters respectively; S522, Prioritization and Personalized Recommendation: Based on Comprehensive Scoring Sort the candidate recommendation items and calculate the recommendation probability by combining Softmax normalization , expressed as: in, is the final recommendation probability of the candidate recommendation item, is the temperature parameter, controlling the degree of differentiation between the recommendations. When it is larger, the probability of highly rated recommended items is higher, and popular recommended items are concentrated. When it is smaller, the recommendation results are more uniform, increasing the possibility of long-tail recommendations. Candidate recommendation The overall score of S523, Personalized recommendation scheme generation: based on the calculated recommendation probability , using the Top-K sampling strategy to generate the final personalized recommendation solution , expressed as: ; in, is the final set of recommended items, is the number of recommended items, such as Representative recommends the top 5 best items; Through the above content, combined with the real-time demand weight of the user portrait, the candidate recommendation items are dynamically scored and prioritized to achieve a personalized, high-precision recommendation strategy. Multi-objective optimization scoring is adopted, and service matching, privacy protection and resource consumption are comprehensively considered to ensure that the recommendation scheme strikes a balance between accuracy, security and computational efficiency. At the same time, the combination of Softmax normalization and Top-K sampling strategy enables the recommendation results to highlight high-matching options while taking into account long-tail recommendations, thereby improving the diversity and adaptability of the recommendation system, thereby providing users with smarter and more accurate Beidou navigation location services.

[0031] like Figure 2 As shown, the Beidou navigation location service recommendation system based on big data is used to implement the above-mentioned Beidou navigation location service recommendation method based on big data, including the following modules: Beidou multi-source data acquisition module: obtains users' Beidou multi-source data through the Beidou satellite navigation system, including real-time positioning data, user historical trajectory data, surrounding POI feature data and meteorological environment data; Dynamic spatiotemporal feature coding module: performs spatiotemporal correlation analysis on Beidou multi-source data and constructs spatiotemporal feature vectors including position precision factor, motion state vector, and environmental feature matrix; Privacy-preserving feature extraction module: It applies differential privacy algorithm to reduce the dimension of spatiotemporal feature vectors and generates a standardized feature data set with privacy protection. User portrait generation module: Based on the standardized feature data set, dynamic label matching is performed in combination with user behavior logs to generate user portraits including real-time demand weights; Multi-objective recommendation generation module: establishes a multi-objective optimization model including service matching degree, privacy protection degree, and resource consumption coefficient, and generates a recommendation plan based on user portraits.

[0032] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0033] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A Beidou navigation location service recommendation method based on big data, characterized in that: The following steps are involved: S1, Beidou multi-source data acquisition: Obtain users' Beidou multi-source data through the Beidou satellite navigation system, including real-time positioning data, user historical trajectory data, surrounding POI feature data and meteorological environment data; S2, dynamic spatiotemporal feature coding: perform spatiotemporal correlation analysis on Beidou multi-source data and construct spatiotemporal feature vectors including position precision factor, motion state vector, and environmental feature matrix; S3, privacy-preserving feature extraction: Apply the differential privacy algorithm to reduce the dimension of the spatiotemporal feature vector to generate a standardized feature dataset with privacy protection; S4, user portrait generation: Based on the standardized feature data set, dynamic label matching is performed in combination with user behavior logs to generate user portraits including real-time demand weights; S5, multi-objective recommendation generation: establish a multi-objective optimization model including service matching degree, privacy protection degree, and resource consumption coefficient, and generate a recommendation plan based on the generated user portrait.

2. The Beidou navigation location service recommendation method based on big data according to claim 1, characterized in that: The Beidou multi-source data acquisition in S1 includes: S11, Beidou real-time positioning data acquisition: through the positioning service of the Beidou satellite navigation system, receive the GNSS signal of the user terminal device to obtain the latitude and longitude coordinates, elevation and speed information of the current time; S12, user historical trajectory data collection: based on the continuous positioning data of the user terminal, the trajectory is stored in combination with the timestamp, and the user's movement mode, stop points and historical travel trajectory are extracted; S13, obtaining surrounding POI feature data: using the user's current location and historical trajectory data, calling the geographic information database or map service interface to obtain surrounding POI information, including transportation hubs, commercial areas, and medical facilities, and correlating the user's visit frequency and preference characteristics; S14, meteorological environment data collection: obtain the weather conditions, temperature and humidity, wind speed, visibility and electromagnetic interference of the user's current location through a third-party meteorological data interface.

3. The Beidou navigation location service recommendation method based on big data according to claim 1, characterized in that: The dynamic spatiotemporal feature encoding in S2 includes: S21, Position DPI calculation: Calculate the user's position DPI based on real-time positioning information from Beidou multi-source data ; S22, motion state vector construction: Based on the user's historical trajectory data, extract the user's speed, acceleration, and heading angle to construct a motion state vector ; S23, environmental feature matrix construction: Based on the meteorological environment data of the user's current location and the surrounding POI information, an environmental feature matrix is ​​constructed. ; S24, spatiotemporal feature vector construction: The position precision factor, motion state vector and environmental feature matrix are integrated to form a spatiotemporal feature vector .

4. The Beidou navigation location service recommendation method based on big data according to claim 3 is characterized in that: The privacy-preserving feature extraction in S3 includes: S31, spatiotemporal feature vector dimensionality reduction preprocessing: principal component analysis is used to reduce the spatiotemporal feature vector Perform dimensionality reduction to obtain the space-time feature vector after dimensionality reduction ; S32, Laplace noise perturbation: spatiotemporal feature vectors after dimensionality reduction Introduce Laplace noise to obtain the noisy spatiotemporal feature vector ; S33, Privacy-preserving feature data standardization: The spatiotemporal feature vector after noise addition Perform normalization to obtain a standardized feature data set .

5. The Beidou navigation location service recommendation method based on big data according to claim 4 is characterized in that: The user portrait generation in S4 includes: S41, User feature vector construction: based on standardized feature dataset and user historical behavior logs, extracting user personalized features and constructing user feature vectors ; S42, Dynamic Tag Matching: Based on User Feature Vector , combined with real-time behavior data, dynamic tag matching, and calculation of tag relevance ; S43, real-time demand weight calculation: tag association based on dynamic matching , calculate the weight of each requirement label ; S44, user portrait update: combine all requirement tags and their weights to generate the final user portrait .

6. The Beidou navigation location service recommendation method based on big data according to claim 5 is characterized in that: The multi-objective recommendation generation in S5 includes: S51, multi-objective optimization model construction: construct a multi-objective optimization model including service matching degree, privacy protection degree and resource consumption coefficient to balance recommendation accuracy, privacy security and resource utilization; S52, personalized recommendation plan generation: Based on the real-time demand weight of the user portrait, the candidate recommendation items are prioritized to generate a personalized recommendation plan.

7. The Beidou navigation location service recommendation method based on big data according to claim 6 is characterized in that: The multi-objective optimization model construction in S51 includes: S511, service matching calculation: Calculate the service matching based on the feature similarity between the user profile and the candidate recommendation items. ; S512, privacy protection calculation: Based on the differential privacy mechanism, calculate the privacy protection degree of the recommended solution for user data ; S513, resource consumption coefficient calculation: comprehensively consider computing cost, bandwidth usage and storage requirements, and calculate the resource consumption coefficient of the recommended solution ; S514, Multi-objective optimization model construction: combined with service matching , privacy protection and resource consumption coefficient , construct the multi-objective optimization objective function.

8. The Beidou navigation location service recommendation method based on big data according to claim 7 is characterized in that: The generation of the personalized recommendation scheme in S52 includes: S521, candidate recommendation item score calculation: Based on the real-time demand weight of the user portrait, the candidate recommendation items are scored and the comprehensive score of the candidate recommendation items is defined. ; S522, Prioritization and Personalized Recommendation: Based on Comprehensive Scoring Sort the candidate recommendation items and calculate the recommendation probability by combining Softmax normalization ; S523, Personalized recommendation scheme generation: based on the calculated recommendation probability , using the Top-K sampling strategy to generate the final personalized recommendation solution .

9. A Beidou navigation location service recommendation system based on big data, used to implement the Beidou navigation location service recommendation method based on big data as described in any one of claims 1 to 8, characterized in that: Includes the following modules: Beidou multi-source data acquisition module: obtains users' Beidou multi-source data through the Beidou satellite navigation system, including real-time positioning data, user historical trajectory data, surrounding POI feature data and meteorological environment data; Dynamic spatiotemporal feature coding module: performs spatiotemporal correlation analysis on Beidou multi-source data and constructs spatiotemporal feature vectors including position precision factor, motion state vector, and environmental feature matrix; Privacy-preserving feature extraction module: It applies differential privacy algorithm to reduce the dimension of spatiotemporal feature vectors and generates a standardized feature data set with privacy protection. User portrait generation module: Based on the standardized feature data set, dynamic label matching is performed in combination with user behavior logs to generate user portraits including real-time demand weights; Multi-objective recommendation generation module: establishes a multi-objective optimization model including service matching degree, privacy protection degree, and resource consumption coefficient, and generates a recommendation plan based on user portraits.

Images