Large model recommendation method and system based on individual user behavior

Abstract

The present invention relates to the field of data processing, and in particular to a large-scale model recommendation method and system based on individual user-side behavior. The method: obtains user behavior data of a target user; inputs the user behavior data into an active analysis model to perform user profiling, and obtains the target user's behavior preference information; based on the behavior preference information and the target user's current geographical location, predicts the target user's behavior trajectory information within a preset time period in the future through a behavior construction model; the behavior trajectory information includes the target user's behavior nodes whose predicted execution probability reaches a preset threshold within a preset time period in the future; and recommends to the target user a comprehensive service unit that matches the current geographical location based on the behavior trajectory information. Through the active analysis model and the behavior construction model, a comprehensive analysis of user behavior and the environment in which they are located is performed from different angles, thereby improving the accuracy and efficiency of information recommendations, providing users with a variety of comprehensive service information, and improving user experience.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a large-scale model recommendation method and system based on individual user-side behavior. Background Art

[0002] User behavior encompasses a wide range of online and offline activities, including shopping, socializing, entertainment, learning, and work. Each behavior has multiple forms and scenarios. For example, users may shop in physical stores or through various e-commerce platforms. The categories, frequency, and timing of purchases vary significantly, making it difficult to accurately describe and predict them using a universal model.

[0003] In related technologies, user behavior encompasses a variety of online and offline activities, including shopping, socializing, entertainment, learning, and work, and each behavior has multiple forms and scenarios. For example, when it comes to shopping, users may shop in physical stores or through different e-commerce platforms. The categories, frequency, and time of purchase vary greatly, making it difficult to accurately describe and predict using a universal model. In addition, with the development of the Internet and mobile devices, user behavior is changing rapidly, requiring models to be able to process and analyze data in real time and update predictions of user behavior in a timely manner. However, some existing models may face problems such as insufficient computing resources and slow processing speeds when processing large-scale real-time data, making them unable to meet the needs of real-time predictions.

[0004] Therefore, it is urgent to design a new technical solution to solve at least one of the above technical problems. Summary of the Invention

[0005] In response to the technical problems existing in the prior art, the present invention provides a large-scale model recommendation method and system based on individual user-side behavior, which is used to solve at least one of the above-mentioned technical problems such as data description differences, insufficient computing resources, and slow processing speed.

[0006] In a first aspect, embodiments of the present application provide a large-scale model recommendation method based on individual user terminal behavior, comprising:

[0007] Obtaining user behavior data of the target user; the user behavior data includes at least: online behavior data and offline behavior data;

[0008] Inputting the user behavior data into an active analysis model to perform user profiling and obtain the behavior preference information of the target user; the active analysis model is used to mine and analyze the behavior preference characteristics in the user behavior data from multiple dimensions;

[0009] Based on the behavior preference information and the current geographic location of the target user, predicting the target user's behavior trajectory information within a preset future time period through a behavior construction model; the behavior trajectory information includes behavior nodes of the target user whose predicted execution probability reaches a preset threshold within the preset future time period;

[0010] Based on the behavior trajectory information, a comprehensive service unit that matches the current geographic location is recommended to the target user; the comprehensive service unit includes a service unit that matches the behavior node in the behavior trajectory information, and the service unit is at least one of offline products, offline stores, offline service businesses, online service businesses, and online products.

[0011] In a second aspect, an embodiment of the present application provides a large-scale model recommendation system based on individual user terminal behavior, the system comprising the following units:

[0012] An acquisition unit, configured to acquire user behavior data of a target user; the user behavior data includes at least online behavior data and offline behavior data;

[0013] An analysis unit is configured to input the user behavior data into an active analysis model to perform user profiling and obtain the behavior preference information of the target user; the active analysis model is configured to mine and analyze the behavior preference features in the user behavior data from multiple dimensions;

[0014] A prediction unit is configured to predict the target user's behavior trajectory information within a preset future time period based on the behavior preference information and the target user's current geographic location through a behavior construction model; the behavior trajectory information includes behavior nodes of which the target user's predicted execution probability reaches a preset threshold within the preset future time period;

[0015] A recommendation unit is used to recommend a comprehensive service unit that matches the current geographical location to the target user based on the behavior trajectory information; the comprehensive service unit includes a service unit that matches the behavior node in the behavior trajectory information, and the service unit is at least one of an offline product, an offline store, an offline service business, an online service business, and an online product.

[0016] In a third aspect, an embodiment of the present application provides an electronic device, comprising:

[0017] at least one processor, memory, and input-output unit;

[0018] The memory is used to store a computer program, and the processor is used to call the computer program stored in the memory to execute the large model recommendation method based on individual user terminal behavior of the first aspect.

[0019] In a fourth aspect, a computer-readable storage medium is provided, which includes instructions. When the instructions are executed on a computer, the computer executes the large model recommendation method based on individual user-side behavior of the first aspect.

[0020] The beneficial effect of the present invention is that it provides a large-scale model recommendation method and system based on individual user-side behavior. In this technical solution, the user behavior data of the target user is obtained; the user behavior data includes at least online behavior data and offline behavior data. Then, the user behavior data is input into the active analysis model to perform user profiling to obtain the behavioral preference information of the target user; the active analysis model is used to mine and analyze the behavioral preference characteristics in the user behavior data from multiple dimensions. Then, based on the behavioral preference information and the current geographical location of the target user, the behavioral trajectory information of the target user within a preset time period in the future is predicted through a behavior construction model; the behavioral trajectory information includes the behavior nodes of the target user whose predicted execution probability reaches a preset threshold within the preset time period in the future. Finally, based on the behavioral trajectory information, a comprehensive service unit that matches the current geographical location is recommended to the target user; the comprehensive service unit includes a service unit that matches the behavior node in the behavioral trajectory information, and the service unit is at least one of offline goods, offline stores, offline service businesses, online service businesses, and online goods.

[0021] The embodiment of the present application, by acquiring user behavior data covering both online and offline, can gain a comprehensive and detailed insight into the user's real behavior patterns and preferences, laying a solid foundation for subsequent precise analysis. The active analysis model deeply mines and analyzes behavioral preference characteristics from multiple dimensions, can accurately construct user portraits, outline user behavior preference information, and allow the system to have a more thorough understanding of the user. Based on behavioral preference information and the user's current geographic location, with the help of a behavioral construction model, behavioral trajectory information within a preset time period in the future is predicted, which can proactively predict the user's possible behavior, and by setting a preset threshold for the prediction execution probability, high-probability behavior nodes are screened out, making the prediction more reliable and practical. Finally, based on the behavioral trajectory information, a comprehensive service unit that matches the current geographic location is recommended, which includes various service units that match the behavior nodes, achieving a high degree of fit between service recommendations and the user's immediate needs and the scenario they are in, greatly improving the accuracy and relevance of recommendations, providing users with services that are more in line with actual needs, enhancing user experience, and also helping to improve the operational efficiency and resource utilization of service providers. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flowchart of a large-scale model recommendation method based on individual user terminal behavior according to an embodiment of the present application;

[0023] Figure 2This is a schematic diagram of the structure of a large-scale model recommendation system based on individual user terminal behavior according to an embodiment of the present application;

[0024] Figure 3 This is a schematic structural diagram of an electronic device according to an embodiment of the present application;

[0025] Figure 4 It is a structural diagram of a medium device according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0027] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the described features. In the description of this application, "plurality" means two or more, unless otherwise specifically specified.

[0028] In the description of this application, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art will recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in this application.

[0029] The embodiment of the present application provides a large-scale model recommendation method and system based on individual user-side behavior. In this technical solution, the user behavior data of the target user is obtained; the user behavior data includes at least online behavior data and offline behavior data. Then, the user behavior data is input into the active analysis model to perform user profiling, and the behavior preference information of the target user is obtained; the active analysis model is used to mine and analyze the behavior preference characteristics in the user behavior data from multiple dimensions. Then, based on the behavior preference information and the current geographical location of the target user, the behavior trajectory information of the target user within a preset time period in the future is predicted through a behavior construction model; the behavior trajectory information includes the behavior nodes of the target user whose predicted execution probability reaches a preset threshold within the preset time period in the future. Finally, based on the behavior trajectory information, a comprehensive service unit that matches the current geographical location is recommended to the target user; the comprehensive service unit includes a service unit that matches the behavior node in the behavior trajectory information, and the service unit is at least one of offline goods, offline stores, offline service businesses, online service businesses, and online goods.

[0030] The embodiment of the present application, by acquiring user behavior data covering both online and offline, can gain a comprehensive and detailed insight into the user's real behavior patterns and preferences, laying a solid foundation for subsequent precise analysis. The active analysis model deeply mines and analyzes behavioral preference characteristics from multiple dimensions, can accurately construct user portraits, outline user behavior preference information, and allow the system to have a more thorough understanding of the user. Based on behavioral preference information and the user's current geographic location, with the help of a behavioral construction model, behavioral trajectory information within a preset time period in the future is predicted, which can proactively predict the user's possible behavior, and by setting a preset threshold for the prediction execution probability, high-probability behavior nodes are screened out, making the prediction more reliable and practical. Finally, based on the behavioral trajectory information, a comprehensive service unit that matches the current geographic location is recommended, which includes various service units that match the behavior nodes, achieving a high degree of fit between service recommendations and the user's immediate needs and the scenario they are in, greatly improving the accuracy and relevance of recommendations, providing users with services that are more in line with actual needs, enhancing user experience, and also helping to improve the operational efficiency and resource utilization of service providers.

[0031] It is understandable that the above-mentioned large-scale model recommendation method and system based on individual user terminal behavior solves the problems existing in the related art in the following ways:

[0032] To address the problem of the difficulty of universal description and prediction of user behavior diversity, the embodiments of the present application obtain the online and offline behavior data of the target user, comprehensively covering the user's shopping, social, entertainment, learning, work and other activity data. For example, not only online data such as shopping records on e-commerce platforms are collected, but also offline data such as consumption records in physical stores. This multi-source data integration method avoids the one-sided information caused by relying on a single data source, thereby gaining a more comprehensive understanding of the overall picture of user behavior and providing a rich data foundation for accurately describing and predicting user behavior.

[0033] Furthermore, active analysis models are used to mine and analyze behavioral preference characteristics within user behavior data from multiple dimensions. Shopping behavior is analyzed across multiple dimensions, including shopping platform, category, frequency, and time. For example, by analyzing shopping frequency across different e-commerce platforms and differences in shopping categories over different time periods, a detailed profile of user shopping behavior preferences is constructed. This multi-dimensional analysis breaks the limitations of a single, general model and more accurately captures user behavior across different manifestations and scenarios, thereby improving prediction accuracy.

[0034] In order to solve the computing resource and speed problems faced by real-time prediction, the embodiment of the present application is based on a large-scale model recommendation system of individual user-side behavior. With the powerful computing and processing capabilities of the large model, it can better cope with large-scale real-time data. The large model has a large number of parameters and strong learning capabilities, which can efficiently process and analyze a large amount of real-time data. Compared with traditional models, it has significant advantages in processing scale and speed. For example, when faced with a large amount of online and offline behavior data generated by users in real time, the large model can extract features and perform analysis more quickly to provide support for real-time prediction.

[0035] In the embodiment of this application, when designing the behavior construction model, the architecture and algorithm are optimized to meet real-time requirements. For example, a stream computing architecture suitable for processing real-time data is adopted, combined with a lightweight but efficient algorithm, which can quickly process data with limited computing resources. When predicting the user's future behavior trajectory information within a preset time period, the algorithm is optimized to reduce computational complexity and increase processing speed, allowing the model to promptly update its prediction of user behavior to meet real-time requirements.

[0036] The large model recommendation scheme based on personal user-side behavior provided in the embodiment of the present application can also be executed by an electronic device, which can be a server, a server cluster, or a cloud server. The electronic device can also be a terminal device such as a mobile phone, a computer, a tablet computer, a wearable device, or a dedicated device (such as a dedicated terminal device with a large model recommendation system based on personal user-side behavior). These electronic devices can also be equipped with the chips introduced in the above embodiments. Alternatively, these electronic devices can also be installed with a service program for executing a large model recommendation scheme based on personal user-side behavior.

[0037] Figure 1 A schematic diagram of a large model recommendation method based on individual user terminal behavior provided in an embodiment of the present application is shown as follows: Figure 1 As shown, the method includes the following steps:

[0038] 101. Obtain user behavior data of target users;

[0039] 102. Input the user behavior data into an active analysis model to perform user profiling and obtain the behavior preference information of the target user;

[0040] 103. Based on the behavior preference information and the current geographic location of the target user, predict the behavior trajectory information of the target user within a preset time period in the future through a behavior construction model;

[0041] 104. Recommending a comprehensive service unit that matches the current geographical location to the target user based on the behavior trajectory information.

[0042] In an embodiment of the present application, the user behavior data includes at least: online behavior data and offline behavior data. Online behavior data includes the user's operations on various network platforms, such as web browsing records, which can reflect their interest tendencies; search history, which can provide insights into their needs and concerns. Online transaction information presents consumption preferences and capabilities, etc. Offline behavior data involves the user's activities in real-life scenarios, such as physical store consumption records, which show actual consumption choices. Participation in offline activities reflects social, hobbies and other aspects. Integrating these two types of data can comprehensively outline the overall picture of user behavior, provide a rich and multi-dimensional information foundation for subsequent analysis, portrait construction and precise recommendations based on user behavior, and help to gain a deeper understanding of user needs and behavior patterns.

[0043] In 101, obtaining target users' behavior data is the foundation for accurate analysis and service. This data can be collected through various channels, covering multiple online and offline areas.

[0044] Online, various internet platforms are important sources of data. For example, e-commerce platforms can record users' browsing histories, providing insights into product categories they're interested in and how long they spend browsing, reflecting their level of interest. Search records clearly demonstrate users' immediate needs and areas of exploration. Transaction records provide even more detailed insights into users' consumption preferences, including the types of goods purchased, brands, price ranges, and frequency of purchases. These data, taken together, paint a clear picture of a user's online consumption behavior. Social platforms can also provide a wealth of information. Users' interactions with friends, posting content, and participating in discussion threads reveal their social circles, interests, hobbies, and values.

[0045] Offline, consumption records at physical stores can reveal users' real-world consumption choices. For example, supermarket shopping lists reflect daily consumption needs, while purchase records at specialty stores reveal preferences for specific brands or categories. Furthermore, participation in offline activities is crucial data; users' participation in exhibitions, lectures, sporting events, and other events can reveal their interests, hobbies, and lifestyles. Smart devices such as mobile payment terminals and phones with positioning capabilities can also capture users' geographic locations. Combined with time data, this can be used to analyze users' travel patterns and frequented locations.

[0046] In this way, the user behavior data collected from multiple online and offline channels can comprehensively and three-dimensionally present the overall behavior of the target users, provide a solid data foundation for the subsequent construction of user portraits, analysis of behavioral preferences, and accurate recommendation services, and help enterprises and service providers better meet user needs and improve service quality and user experience.

[0047] In an embodiment of the present application, the active analysis model is used to mine and analyze the behavioral preference characteristics in the user behavior data from multiple dimensions.

[0048] For example, the principle of an active analysis model can be based on deep learning and data mining techniques, using algorithms such as neural networks and decision trees to construct a complex model structure. The model first preprocesses the acquired user behavior data, unifying the format and cleaning up the noisy data. It then performs feature extraction on the data, converting the raw data into a vector form that can be understood by the model. Next, the model's multi-layered structure automatically learns hidden patterns and relationships in the data. For example, convolutional layers are used to capture local features in the data, and recurrent layers are used to process sequential data to analyze the time series characteristics of behavior, thereby exploring behavioral preference characteristics from multiple dimensions such as consumption, social interaction, and interests.

[0049] Therefore, through active analysis models, we can gain deep insights into the real preferences behind user behavior, accurately extract users' interests and behavioral tendencies in different fields, and provide accurate basis for user portraits. The recommendation system can then provide users with highly personalized and demand-oriented recommendation content based on these accurate preference characteristics, effectively improving users' acceptance and satisfaction with recommended content, and enhancing the interactive stickiness between users and the system.

[0050] In the embodiment of the present application, the behavior trajectory information includes the behavior nodes of the target user whose predicted execution probability reaches a preset threshold within a preset time period in the future. Specifically, in the embodiment of the present application, the behavior trajectory information is crucial to accurately grasp the future behavior of the target user. The behavior nodes included therein are derived based on the prediction of the target user's behavior. These predictions are not random guesses, but are scientifically calculated through the behavior construction model based on multiple factors such as user behavior preference information and current geographical location. The predicted execution probability reflects the possibility of the behavior node actually occurring within a preset time period in the future. The preset threshold is an artificially set standard value. Only behavior nodes whose predicted execution probability reaches this threshold will be included in the behavior trajectory information. This setting effectively filters out behavior predictions with lower probability, ensures that the behavior trajectory information focuses on behaviors with high probability of occurrence, and makes the comprehensive service units recommended to users based on this information more targeted and practical, better fits the real needs of users, and improves the accuracy and effectiveness of recommendations.

[0051] In an embodiment of the present application, the comprehensive service unit includes a service unit that matches the behavior node in the behavior trajectory information, and the service unit is at least one of an offline product, an offline store, an offline service business, an online service business, and an online product.

[0052] For example, if the target user's behavior trajectory information predicts that they have a high probability of purchasing sports equipment in a nearby shopping mall within the next hour (behavior node), the comprehensive service unit will match this behavior. This may include offline products, such as sports shoes and sportswear of a certain brand in the mall; offline stores, such as well-known sports brand stores; offline service businesses, such as on-site try-on and customization services for sports equipment. If the user is in an online scene, the service unit may also have online service businesses, such as online push of sports and fitness courses; online products, such as the same sports equipment recommended online by sports apps.

[0053] Clearly, this matching approach can greatly improve the accuracy of recommendations, precisely targeting users' potential needs. For users, this allows them to quickly access services and products that interest them and are relevant to their current context, enhancing the user experience and increasing their attention and engagement with recommended content. For businesses, precise recommendations help increase product and service conversion rates, optimize resource allocation, reduce ineffective promotion costs, improve operational efficiency and commercial benefits, and promote the coordinated development of online and offline businesses.

[0054] As an optional embodiment, the active analysis model includes at least the following structures: an extraction layer, a construction layer, an analysis layer, a prediction layer, and an output layer.

[0055] Based on the above model structure, in step 102, the user behavior data is input into the active analysis model to perform user profiling, and the behavior preference information of the target user is obtained, including:

[0056] Extracting multi-dimensional user behavior features from the user behavior data through an extraction layer;

[0057] Through the construction layer, the multidimensional user behavior features are projected into the multidimensional portrait space according to the projection strategy corresponding to each dimension to obtain the corresponding multidimensional user portrait; through the analysis layer, the multidimensional user portrait is adaptively segmented using pre-configured slice fitting parameters to obtain multiple behavior slices corresponding to the multidimensional user portrait; through the prediction layer, the matching probability between multiple behavior slices and preset behavior types is predicted to select candidate behavior types for the multidimensional user portrait; through the output layer, the behavior preference information of the target user is generated according to the candidate behavior types.

[0058] In an embodiment of the present application, each portrait coordinate point in the multi-dimensional user portrait corresponds to at least one behavior feature point in the multi-dimensional user behavior feature.

[0059] In the process of constructing a multidimensional user profile, the multidimensional user profile is like a complex coordinate space, where each of its coordinate points has special significance. These coordinate points do not exist in isolation, but are closely linked to the multidimensional user behavior characteristics. Each profile coordinate point corresponds to at least one behavioral feature point in the multidimensional user behavior characteristics. For example, in a multidimensional user profile space constructed based on user consumption frequency, consumption amount, and consumption category, a specific profile coordinate point may correspond to both a user's high-frequency consumption behavior feature point in the electronic product category and a behavioral feature point associated with the user's high-amount consumption in the same category. This correspondence enables the multidimensional user profile to accurately map the user's complex behavioral characteristics. By analyzing these coordinate points, we can gain a deep understanding of the user's behavioral patterns across different dimensions, providing a rich and accurate information foundation for further user behavior analysis, personalized recommendations, and other applications.

[0060] In an embodiment of the present application, multiple behavioral slices are used to represent the behavioral change characteristics of the multi-dimensional user portrait under different analysis angles; the slice fitting parameters are used to represent the slicing method of the multi-dimensional user portrait and the method of selecting the slice cutting position.

[0061] It's understandable that multiple behavior slices are used to represent the behavioral changes of a multi-dimensional user profile from different analytical perspectives. These behavior slices divide the user profile into multiple parts, facilitating analysis of user behavior from multiple dimensions and perspectives. For example, in a three-dimensional user profile, one behavior slice might represent a specific user behavior pattern in a certain time period and scenario, while another behavior slice might represent a different user behavior pattern under different conditions.

[0062] By analyzing different behavioral slices, we can gain a more detailed understanding of user behavioral preferences, providing richer information for subsequent prediction layers. Different behavioral slices can reflect user behavioral tendencies in different aspects. For example, in a shopping scenario, one slice may reflect the user's behavioral characteristics when purchasing high-value items, while another slice may reflect the user's behavioral characteristics when purchasing daily necessities.

[0063] Specifically, in step 102, the extraction layer acts as a data filter, selecting representative multidimensional user behavior features from complex user behavior data. For example, it extracts features such as spending amount and interaction frequency from various behavioral data such as shopping and social interaction. Next, the construction layer projects these features into a multidimensional profile space based on the projection strategies corresponding to each dimension. This is like placing different information points in a multidimensional coordinate system, forming a multidimensional user profile that comprehensively outlines the user's behavior. The analysis layer then uses pre-configured slice fitting parameters to adaptively segment the multidimensional user profile based on data characteristics and analysis requirements, generating multiple behavioral slices that reflect user behavior changes from different perspectives. Based on this, the prediction layer evaluates the probability of these behavioral slices matching pre-set behavior types. For example, it determines the compatibility of the behavior pattern represented by a behavior slice with pre-set types such as "sports enthusiast" or "fashion seeker," and selects those with high matching probabilities as candidate behavior types. Finally, the output layer generates behavioral preference information for the target user based on the candidate behavior types, completing the key step of user profiling.

[0064] For example, assume user behavior data covers online shopping, social platform interactions, and other information. The extraction layer extracts multidimensional features from this data, such as the user's monthly online clothing purchase amount and the frequency of posting in fitness-related social groups. The construction layer projects these features into a multidimensional portrait space composed of consumption and social dimensions to form a user profile. The analysis layer slices the data according to preset parameters, such as spending ranges and social activity levels, to generate multiple behavioral slices. The prediction layer determines the probability of these slices matching preset behavioral types such as "fashion consumers" and "fitness enthusiasts." If a slice has a high probability of matching "fitness enthusiast," it is selected as a candidate type. Based on this, the output layer generates behavioral preference information for the user's preferences for fitness-related activities and products.

[0065] This approach enables deep mining of user behavior data, accurate user profiles, and a comprehensive and detailed understanding of user behavior preferences. This generated behavioral preference information provides strong support for applications such as personalized recommendations and precision marketing, improving the accuracy and relevance of recommendations and enhancing the user experience.

[0066] Further optionally, in the above steps, the prediction layer is used to predict the matching probability between multiple behavior slices and preset behavior types to select candidate behavior types for the multi-dimensional user portrait, including:

[0067] For the j-th preset behavior type Tj among multiple preset behavior types, calculate the matching degree P(Tj) between multiple behavior slices and the preset behavior type Tj; summarize the matching degrees between the multiple preset behavior types and the preset behavior type Tj to obtain the matching probability P(Tj) between the multiple preset behavior types and the preset behavior type Tj; select the candidate behavior type whose matching probability meets the preset screening conditions from the multiple preset behavior types.

[0068] Among them, P(Tj)i between the i-th behavior slice Si and the preset behavior type Tj is expressed as the following formula: cosine(Si,Tj) is the cosine similarity between the behavior slice Si and the preset behavior type Tj, N(Si,Tj) is the matching probability vector obtained by processing the behavior slice Si and the preset behavior type Tj through the preset neural network, n is the number of behavior slices, and wi is the weight coefficient corresponding to the behavior slice Si.

[0069] During the prediction process, this method comprehensively considers multiple matching metrics, leveraging the intuitiveness of cosine similarity and the ability of neural networks to uncover latent relationships. It flexibly adjusts the contribution of different behavior slices to matching through weight coefficients, making matching probability calculation more comprehensive and accurate. This helps accurately identify candidate behavior types that are highly correlated with user behavior characteristics, laying the foundation for generating accurate user behavior preference information. This improves the effectiveness and accuracy of user-profile-based recommendation systems, marketing campaigns, and other applications, better meeting user personalization needs, and enhancing user experience and business efficiency.

[0070] Further optionally, in the above steps, the matching probability P(Tj) between the plurality of preset behavior types and the preset behavior type Tj is expressed as the following formula:

[0071]

[0072] Among them, μi and μj are the means of the behavior slice Si and another behavior slice Sj respectively, is an exponential term based on the covariance matrix, which is used to represent the correlation between a behavior slice Si and another behavior slice Sj.

[0073] This formula further considers the correlation between behavior slices when calculating the matching probability between multiple preset behavior types and preset behavior types. First, the cosine similarity between the behavior slice and the preset behavior type and the matching probability vector obtained through the preset neural network processing are combined with the weight coefficient corresponding to the behavior slice to obtain the preliminary matching degree between the behavior slice and the preset behavior type. On this basis, an exponential term based on the covariance matrix is introduced. The covariance matrix can reflect the relationship between behavior slices. The sum is the mean of the behavior slice and the mean of another behavior slice respectively. By calculating the exponential term, the degree of correlation between different behavior slices is measured. Behavior slices with high correlation have a more significant impact on the matching probability. The exponential term incorporates this correlation into the calculation of the matching probability, so that the final matching probability not only considers the matching situation of a single behavior slice and the preset behavior type, but also takes into account the internal connection between the behavior slices, thereby more comprehensively and accurately reflecting the degree of fit between the user's behavior characteristics and the preset behavior type.

[0074] This method of calculating matching probabilities allows for a more refined characterization of user behavior patterns. By considering the correlations between behavior slices, we avoid viewing each behavior slice in isolation, uncover the underlying logical relationships between user behaviors, and improve the accuracy of matching probabilities. Selecting candidate behavior types based on more accurate matching probabilities helps build more precise user profiles, enabling subsequent recommendations and analysis based on user profiles to better meet actual user needs, providing users with more personalized, high-quality services and enhancing the system's competitiveness in the field of user behavior analysis.

[0075] Further optionally, in the above steps, the multi-dimensional user profile is adaptively segmented by using pre-configured slice fitting parameters through the analysis layer to obtain multiple behavior slices corresponding to the multi-dimensional user profile, including:

[0076] Based on the method of selecting the slice interception position, the starting position and the ending position of each behavior slice are set; the computational complexity corresponding to each behavior slice under different analysis angle combinations is obtained, and the target analysis angle combination matching each behavior slice is selected based on the obtained computational complexity; based on the starting position and the ending position of each behavior slice, the multi-dimensional user portrait is adaptively segmented from the slice perspective corresponding to the target analysis angle combination to obtain multiple behavior slices; wherein the adaptive segmentation is at least one of an equidistant division method, a peak division based on data distribution, and a cluster division.

[0077] Specifically, first, based on the pre-defined slice cutout location selection method, the starting and ending positions of each behavior slice in the multidimensional user profile space are determined. This setting forms the basis for segmentation and determines the specific range of the behavior slice in space. Next, the computational complexity of each behavior slice is calculated for different analysis angle combinations. Different analysis angle combinations have different data processing and analysis requirements, and the computational complexity will also vary. By evaluating the computational complexity, we can understand the difficulty and resource consumption of different analysis angle combinations when processing specific behavior slices. Then, based on the computational complexity, the most suitable target analysis angle combination is selected for each behavior slice, ensuring that the subsequent analysis can effectively explore the characteristics of the behavior slice while achieving a balance between computing resources and efficiency. Finally, starting from the slice perspective corresponding to the selected target analysis angle combination, adaptive segmentation methods such as equidistant segmentation, peak segmentation based on data distribution, or cluster segmentation are used to segment the multidimensional user profile to obtain multiple behavior slices. Equal-interval partitioning divides data at fixed intervals and is suitable for situations where data distribution is relatively uniform. Peak partitioning based on data distribution divides data at peaks and divides data at valleys according to the density of data distribution, which can better reflect the actual distribution of data. Cluster partitioning groups data points with similar behavioral characteristics into one category to form slices, highlighting the differences in behavioral patterns.

[0078] It is understandable that the following is a detailed introduction to the equal distance division method, peak division based on data distribution, and cluster division:

[0079] Equidistant partitioning: Each dimension of the multidimensional user profile space is divided into fixed lengths or intervals. In two-dimensional space, for example, fixed partition lengths are set on both the x-axis and the y-axis, cutting the entire space into small, equally sized regions. Each small region is a behavioral slice. This partitioning method does not consider the specific distribution of the data and operates entirely on the principle of equidistant spacing, ensuring that each slice has the same length or range in each dimension.

[0080] This method is simple and intuitive, easy to understand and implement. The partitioning process does not require complex statistical analysis or calculations on the data. It can quickly segment the multidimensional space, obtaining a defined number of uniformly sized behavioral slices, facilitating subsequent unified processing and analysis of each slice. This approach is suitable for scenarios where data is relatively evenly distributed across all dimensions, or where there is no prior knowledge of the data distribution and a desire to partition the space in a simple, unified manner.

[0081] Peak segmentation based on data distribution: First, statistical analysis is performed on each dimension of the user profile data, using methods such as kernel density estimation to obtain the probability density function for that dimension. By analyzing the probability density function, the peaks and valleys of the data distribution are identified. At peaks, the data distribution is dense, indicating that user behavior is concentrated in these areas. Therefore, these areas are divided into finer slices to more accurately capture user behavior characteristics. At valleys, the data distribution is relatively sparse, indicating relatively less user behavior, so they are divided into coarser slices. In this way, different segmentation granularities are used at different locations based on the actual data distribution, so that the slices can more accurately reflect the concentration of user behavior in different areas.

[0082] In this way, the data can be flexibly divided according to its actual distribution, providing finer slices in data-dense areas and capturing user behavior characteristics more comprehensively. In data-sparse areas, the division granularity is appropriately relaxed to avoid excessive and meaningless divisions in areas with small amounts of data, thereby improving the efficiency and accuracy of the division and making the resulting behavior slices better reflect the true distribution of user behavior.

[0083] Therefore, this method is suitable for scenarios where data distribution is uneven and it is hoped that a more detailed description of the concentration of user behavior in different areas can be achieved. For example, the distribution of user behavior in different time periods, different consumption amount ranges, and other dimensions can be analyzed.

[0084] Clustering: Clustering algorithms, such as K-Means, are used to process multidimensional user profile data. Data points are divided into clusters based on metrics such as similarity or distance. Each cluster can be considered a behavioral slice, with its boundaries determined by the range of data points within the cluster. Clustering algorithms automatically group data points with similar behavioral characteristics into the same cluster, resulting in distinct behavioral differences between data points in different clusters, forming distinct behavioral slices that effectively reflect different user behavior patterns.

[0085] This automatically discovers the inherent structure and patterns in the data, clustering data points with similar behavioral characteristics. The resulting behavioral slices accurately reflect the diverse behavioral patterns of users, helping to gain a deeper understanding of the diversity and differences in user behavior. The slices' location and shape don't need to be determined in advance; they are segmented entirely based on the characteristics of the data itself, making them highly adaptable.

[0086] Therefore, it is suitable for scenarios where it is necessary to explore different user behavior patterns, and the data volume is large and the data structure is relatively complex. For example, the comprehensive behavioral characteristics of users can be analyzed and users can be divided into different behavioral groups for targeted marketing or services.

[0087] Obviously, this adaptive segmentation method is highly flexible and targeted. By reasonably setting the slice position, it is ensured that each behavioral slice can effectively cover valuable user behavior characteristics. The target analysis angle combination is selected based on computational complexity to avoid resource waste or poor analysis results caused by improper analysis angles, thereby improving analysis efficiency and accuracy. By adopting a variety of adaptive segmentation methods, the most appropriate division method can be selected according to the characteristics of multi-dimensional user portrait data, so that the obtained behavioral slices can more accurately reflect the characteristics of user behavior in different dimensions and scenarios. Whether it is evenly distributed data, data with obvious peaks, or data based on behavioral pattern clustering, it can be effectively processed. These behavioral slices provide a high-quality data foundation for subsequent in-depth understanding of user behavior, accurate prediction, and personalized recommendations, enhancing the comprehensiveness and effectiveness of the entire system's user behavior analysis.

[0088] As an optional embodiment, in step 103, based on the behavior preference information and the current geographic location of the target user, the behavior trajectory information of the target user within a preset time period in the future is predicted by building a behavior model, including:

[0089] Get the target user's current geographic location;

[0090] Based on the current geographic location and the mapping relationship between the geographic location type and the duration configuration strategy, set the future preset duration corresponding to the target user;

[0091] Obtain corresponding surrounding environment data based on the current geographical location;

[0092] Based on the configured future preset duration, the behavior preference information, the surrounding environment data, and the current geographical location are input into the behavior construction model to perform behavior trajectory prediction to obtain the behavior trajectory information.

[0093] Specifically, the target user's current geographic location must first be determined, forming the basis for subsequent analysis. Next, based on the mapping between geographic location type and duration configuration strategy, a future preset duration tailored to the target user is set. For example, if the user is in a shopping mall, a shorter preset duration might be configured, given the typical duration of activities within the mall; whereas if the user is at an airport, a longer preset duration might be configured due to activities such as waiting for their flight. This allows for more rational planning of the forecast timeframe based on common behavioral duration patterns across different scenarios.

[0094] Next, we obtain the surrounding environment data corresponding to the target user's current location, such as the type of surrounding venues, the flow of people, etc. This environmental data has a significant impact on user behavior. For example, if there is a movie theater nearby, the user may be inclined to watch a movie.

[0095] Finally, the configured future preset duration is input into the behavior model, along with behavioral preference information, surrounding environment data, and current location. The model integrates this information and, using its algorithms and mechanisms, predicts the target user's behavior trajectory within the preset future duration, generating behavioral trajectory information. This approach comprehensively considers factors such as the user's location, environment, and preferences, making the predicted behavioral trajectory information more accurate and providing a precise basis for subsequent service recommendations.

[0096] As an optional embodiment, in step 103, it is assumed that the behavior construction model includes at least the following structures: an extraction layer, a fusion layer, a prediction layer, and a screening layer. Based on this, in step 104, based on the configured future preset time, the behavior preference information, the surrounding environment data, and the current geographical location are input into the behavior construction model to perform behavior trajectory prediction, and the behavior trajectory information obtained includes:

[0097] Through the extraction layer, behavioral preference features are extracted from the behavioral preference information, and surrounding environment features are extracted from the surrounding environment data; through the fusion layer, the behavioral preference features, the surrounding environment features and the current geographical location are associated and fused to obtain the fused behavioral features of the target user; the fused behavioral features include each behavioral preference feature associated with multiple combinations of different surrounding environment features and different geographical locations; through the prediction layer, a multi-branch decision layer is used to predict the changes of the fused behavioral features in the future preset time period, and the predicted behavior nodes of the target user in the future preset time period and the corresponding predicted execution probability are obtained; through the screening layer, the predicted behavior nodes with the preset threshold of the predicted execution probability are constructed as the behavioral trajectory information.

[0098] Extraction layer: This layer is primarily responsible for feature extraction. It accurately extracts behavioral preference features from behavioral preference information. These features cover a user's various preferences, such as shopping, entertainment, and social interactions. It also extracts environmental features from surrounding data, including information about surrounding facilities, pedestrian flow, and environmental type. The purpose of this layer is to transform raw data into more representative and analyzable feature data, providing a foundation for subsequent processing.

[0099] Fusion layer: This layer correlates and fuses the extracted behavioral preference features, surrounding environment features, and current geographic location. Through a specific fusion strategy, different types of information are organically combined to form a fused behavioral profile for the target user. For example, a user's shopping preferences, information about nearby shopping malls, and current location information are combined into a fused behavioral profile. This includes multiple different combinations, allowing each behavioral preference feature to be associated with different surrounding environment features and different geographic locations, reflecting the comprehensive behavioral characteristics that users may exhibit in different environments and locations.

[0100] Prediction Layer: This layer uses a multi-branch decision-making layer to predict how the fused behavioral features will evolve over a preset future timeframe. This layer can be understood as having multiple decision branches, each tailored to a specific scenario, taking into account multiple possibilities and conditions. By analyzing the fused behavioral features, it predicts various possible behavior nodes that the target user may experience over a preset future timeframe, along with the predicted execution probability for each behavior node. Leveraging its powerful decision-making capabilities, this layer assesses user behavior trends under the influence of different environments and preferences, determining the likelihood of each behavior occurring.

[0101] Screening layer: This layer serves as the final checkpoint, filtering the prediction results. Predicted behavior nodes whose predicted execution probability does not reach the preset threshold are excluded, retaining only those that meet or exceed the threshold and constructing behavioral trajectory information. This ensures the reliability and effectiveness of the behavioral trajectory information and avoids interference from excessive low-probability behaviors.

[0102] Based on the above structure, the extraction layer first applies data processing and feature extraction techniques, potentially employing methods such as data cleaning, feature encoding, and feature selection, to transform raw behavioral preference information and surrounding environment data into more valuable behavioral preference and surrounding environment features. For example, a user's shopping history can be converted into a user's preference profile for different product categories, and surrounding environment information such as store type and distance can be extracted as environmental features. Next, the fusion layer fuses these features, using methods such as feature concatenation and weighted fusion to deeply integrate behavioral preference features, surrounding environment features, and current location information. For example, a user's preferred sports category can be combined with information about sports venues near the current location to form a fused behavioral profile, indicating the user's potential for engaging in related sports near that location. The prediction layer then leverages the complex structure of the multi-branch decision layer, employing machine learning or deep learning algorithms to predict the user's behavior within a preset timeframe based on historical data and established behavioral patterns. For the fused behavioral profile, the multi-branch decision layer can develop different prediction paths based on different feature combinations and conditions, calculating the predicted execution probability of each possible behavior node. For example, based on a user's shopping preferences, nearby shopping mall information, and location, the probability of their entering a specific store within a preset timeframe is predicted. Finally, at the screening layer, based on a preset threshold, simple probability comparisons are performed to select the most likely behavior nodes. For example, if the threshold is set to 60%, behavior nodes with a predicted execution probability greater than or equal to 60% are retained, forming the final behavior trajectory information for subsequent recommendation or service decision-making.

[0103] Through this model structure and implementation method, we can analyze user behavior information from multiple perspectives, comprehensively consider user preferences, environment and location factors, and more accurately predict user behavior trajectories in the future, providing users with more targeted and practical services or recommendations.

[0104] Alternatively, the multi-branch decision layer predicts the fusion behavior characteristics, which is a relatively complex but orderly process. The following is its working principle and method:

[0105] First, the multi-branch decision layer takes the fused behavioral features as input. These fused behavioral features include the target user's behavioral preference characteristics, surrounding environment characteristics, current geographical location and other information, and are a comprehensive and rich feature set.

[0106] The multi-branch decision layer then uses its pre-trained multiple branch structures to analyze and process different dimensions or directions based on these input features. Each branch may be responsible for processing a specific type of information or making predictions about specific behavioral patterns. For example, one branch may focus on time-related features to analyze possible behavioral trends over a preset period of time in the future; another branch may focus on location-related information to predict possible behavioral nodes in different geographic locations; and yet another branch may focus on the interaction between behavioral preferences and the surrounding environment to determine the more likely behaviors based on user preferences in a specific environment.

[0107] Each branch typically contains multiple neurons or processing units, which perform weighted summation and nonlinear transformations on the input fused behavioral features to extract higher-level feature representations. These operations are performed based on pre-trained weight parameters, which are continuously adjusted and learned through extensive training data and optimization algorithms, enabling each branch to accurately capture the characteristic patterns relevant to its domain.

[0108] Each branch then generates preliminary predictions based on its own processing results. These predictions may include predicted behavior nodes and corresponding probability distributions under different assumptions or conditions. For example, one branch might predict that a user has a certain probability of appearing at a specific location and performing a specific behavior within a certain time period in the future, while another branch might predict a different probability of performing a different behavior.

[0109] Finally, the multi-branch decision layer integrates and comprehensively evaluates the prediction results from each branch. This may involve weighted fusion and probabilistic combination of predictions from different branches to arrive at a final, multi-faceted prediction result: the target user's predicted behavior node at a preset time in the future and the corresponding predicted execution probability. This comprehensive result, derived from in-depth analysis and prediction of the integrated behavioral features from multiple branches from different perspectives, more comprehensively and accurately reflects the target user's likely future behavioral trajectory and trends.

[0110] In 104 , based on the behavior trajectory information, a comprehensive service unit matching the current geographical location is recommended to the target user.

[0111] In an optional embodiment of step 104, operations are first performed based on previously predicted behavioral trajectory information. This behavioral trajectory information includes the target user's predicted behavior nodes within a preset future timeframe whose execution probability reaches a preset threshold. These nodes reflect the user's likely behavioral trajectory. Simultaneously, the system combines the target user's current geographic location to provide users with accurate and context-sensitive service recommendations.

[0112] The system combines the behavior nodes in the behavior trajectory information with the actual situation of the current geographic location. For example, if the behavior trajectory information shows that the user may have a behavior node of purchasing sports equipment in the future, and there is a shopping mall near the current location, the system will search for service units matching the behavior node of purchasing sports equipment within the mall.

[0113] The comprehensive service units here include various types, such as offline products, that is, specific sports brand products in shopping malls; offline stores, such as professional sports stores; offline service businesses, such as trying on and customizing sports equipment; if the user is in an online scene, there will also be online service businesses, such as online push of sports and fitness courses; and online products, such as the same sports equipment recommended online by sports apps.

[0114] In this way, the system recommends comprehensive service units to target users based on the matching of behavioral trajectory information and current geographic location, so that the recommended services are consistent with the user's behavioral tendencies and the actual location at the moment, greatly improving the accuracy and practicality of the recommendations, meeting the user's potential needs in actual scenarios, and improving the user experience. At the same time, it also helps to improve the resource utilization efficiency and commercial benefits of the service provider.

[0115] The embodiment of the present application, by acquiring user behavior data covering both online and offline, can gain a comprehensive and detailed insight into the user's real behavior patterns and preferences, laying a solid foundation for subsequent precise analysis. The active analysis model deeply mines and analyzes behavioral preference characteristics from multiple dimensions, can accurately construct user portraits, outline user behavior preference information, and allow the system to have a more thorough understanding of the user. Based on behavioral preference information and the user's current geographic location, with the help of a behavioral construction model, behavioral trajectory information within a preset time period in the future is predicted, which can proactively predict the user's possible behavior, and by setting a preset threshold for the prediction execution probability, high-probability behavior nodes are screened out, making the prediction more reliable and practical. Finally, based on the behavioral trajectory information, a comprehensive service unit that matches the current geographic location is recommended, which includes various service units that match the behavior nodes, achieving a high degree of fit between service recommendations and the user's immediate needs and the scenario they are in, greatly improving the accuracy and relevance of recommendations, providing users with services that are more in line with actual needs, enhancing user experience, and also helping to improve the operational efficiency and resource utilization of service providers.

[0116] Figure 2 This is a structural diagram of a large model recommendation system based on individual user terminal behavior provided in an embodiment of the present application, such as Figure 2 As shown, the system includes the following steps:

[0117] An acquisition unit, configured to acquire user behavior data of a target user; the user behavior data includes at least online behavior data and offline behavior data;

[0118] An analysis unit is configured to input the user behavior data into an active analysis model to perform user profiling and obtain the behavior preference information of the target user; the active analysis model is configured to mine and analyze the behavior preference features in the user behavior data from multiple dimensions;

[0119] A prediction unit is configured to predict the target user's behavior trajectory information within a preset future time period based on the behavior preference information and the target user's current geographic location through a behavior construction model; the behavior trajectory information includes behavior nodes of which the target user's predicted execution probability reaches a preset threshold within the preset future time period;

[0120] A recommendation unit is used to recommend a comprehensive service unit that matches the current geographical location to the target user based on the behavior trajectory information; the comprehensive service unit includes a service unit that matches the behavior node in the behavior trajectory information, and the service unit is at least one of an offline product, an offline store, an offline service business, an online service business, and an online product.

[0121] Further optionally, the active analysis model includes at least the following structures: an extraction layer, a construction layer, an analysis layer, a prediction layer, and an output layer; an analysis unit inputs the user behavior data into the active analysis model to perform user profiling and obtain the behavior preference information of the target user, specifically for:

[0122] Extracting multi-dimensional user behavior features from the user behavior data through an extraction layer;

[0123] By constructing layers, the multidimensional user behavior features are projected into a multidimensional portrait space according to the projection strategies corresponding to the respective dimensions to obtain a corresponding multidimensional user portrait; wherein each portrait coordinate point in the multidimensional user portrait corresponds to at least one behavior feature point in the multidimensional user behavior features;

[0124] The multidimensional user profile is adaptively segmented using pre-configured slice fitting parameters through the analysis layer to obtain multiple behavioral slices corresponding to the multidimensional user profile; wherein the multiple behavioral slices are respectively used to represent the behavioral change characteristics of the multidimensional user profile under different analysis angles; the slice fitting parameters are used to indicate the slicing method of the multidimensional user profile and the method of selecting the slicing positions;

[0125] Through the prediction layer, the matching probability between multiple behavior slices and preset behavior types is predicted to select candidate behavior types for the multi-dimensional user portrait;

[0126] Through the output layer, the behavior preference information of the target user is generated according to the candidate behavior type.

[0127] Further optionally, the analysis unit predicts the matching probability between multiple behavior slices and preset behavior types through the prediction layer to select candidate behavior types for the multi-dimensional user portrait, specifically for:

[0128] For the j-th preset behavior type Tj among the multiple preset behavior types, calculating the matching degree P(Tj) between the multiple behavior slices and the preset behavior type Tj;

[0129] Among them, P(Tj)i between the i-th behavior slice Si and the preset behavior type Tj is expressed as the following formula: cosine(Si,Tj) is the cosine similarity between the behavior slice Si and the preset behavior type Tj, N(Si,Tj) is the matching probability vector obtained by processing the behavior slice Si and the preset behavior type Tj through the preset neural network, n is the number of behavior slices, and wi is the weight coefficient corresponding to the behavior slice Si;

[0130] Summarize the matching degrees between the multiple preset behavior types and the preset behavior type Tj to obtain the matching probability P(Tj) between the multiple preset behavior types and the preset behavior type Tj;

[0131] A candidate behavior type whose matching probability meets the preset screening condition is selected from multiple preset behavior types.

[0132] Further optionally, the matching probability P(Tj) between the plurality of preset behavior types and the preset behavior type Tj is expressed as the following formula:

[0133]

[0134] Among them, μi and μj are the means of the behavior slice Si and another behavior slice Sj respectively, is an exponential term based on the covariance matrix, which is used to represent the correlation between a behavior slice Si and another behavior slice Sj.

[0135] Further optionally, the analysis unit, through the analysis layer, uses pre-configured slice fitting parameters to adaptively segment the multi-dimensional user profile to obtain multiple behavior slices corresponding to the multi-dimensional user profile, specifically for:

[0136] Based on the selection method of the slice interception position, set the starting position and the ending position of each behavior slice;

[0137] Obtain the computational complexity corresponding to each behavior slice under different analysis angle combinations, and select the target analysis angle combination that matches each behavior slice based on the obtained computational complexity;

[0138] Based on the starting position and ending position of each behavior slice, the corresponding slice perspectives are combined from the perspective of target analysis to adaptively segment the multi-dimensional user portrait to obtain multiple behavior slices; wherein the adaptive segmentation is at least one of equidistant segmentation, peak segmentation based on data distribution, and cluster segmentation.

[0139] Further optionally, the prediction unit, based on the behavior preference information and the current geographic location of the target user, predicts the behavior trajectory information of the target user within a preset time period in the future through a behavior construction model, specifically for:

[0140] Get the target user's current geographic location;

[0141] Based on the current geographic location and the mapping relationship between the geographic location type and the duration configuration strategy, set the future preset duration corresponding to the target user;

[0142] Obtain corresponding surrounding environment data based on the current geographical location;

[0143] Based on the configured future preset duration, the behavior preference information, the surrounding environment data, and the current geographical location are input into the behavior construction model to perform behavior trajectory prediction to obtain the behavior trajectory information.

[0144] Further optionally, the behavior construction model includes at least the following structures: an extraction layer, a fusion layer, a prediction layer, and a screening layer; a prediction unit, based on a configured future preset time, inputs the behavior preference information, the surrounding environment data, and the current geographical location into the behavior construction model to perform behavior trajectory prediction, thereby obtaining the behavior trajectory information, specifically for:

[0145] extracting behavior preference features from the behavior preference information and extracting surrounding environment features from the surrounding environment data through an extraction layer;

[0146] The behavior preference characteristics, the surrounding environment characteristics, and the current geographical location are associated and fused through a fusion layer to obtain a fused behavior characteristic of the target user; the fused behavior characteristic includes associations of multiple combinations of each behavior preference characteristic with different surrounding environment characteristics and different geographical locations;

[0147] Through the prediction layer, a multi-branch decision layer is used to predict the changes of the fused behavior features in the future preset time length, and obtain the predicted behavior nodes of the target user in the future preset time length and the corresponding predicted execution probability;

[0148] Through the screening layer, the predicted behavior nodes with the preset threshold of the predicted execution probability are constructed as the behavior trajectory information.

[0149] See also Figure 3 , Figure 3 This is a schematic diagram of an embodiment of an electronic device provided in an embodiment of the present application. Figure 3 As shown, an embodiment of the present application provides an electronic device 500, including a memory 510, a processor 520, and a computer program 511 stored in the memory 510 and executable on the processor 520. When the processor 520 executes the computer program 511, the aforementioned embodiment is implemented.

[0150] See also Figure 4 , Figure 4 This is a schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present application. Figure 4 As shown, this embodiment provides a computer-readable storage medium 600 on which a computer program 611 is stored. When the computer program 611 is executed by a processor, the aforementioned embodiment is implemented.

[0151] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0152] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0153] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks 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 computer, or other programmable data processing device to produce 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 1a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0154] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0155] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0156] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0157] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A large-scale model recommendation method based on individual user behavior, characterized in that: The method at least comprises: Obtaining user behavior data of the target user; the user behavior data includes at least: online behavior data and offline behavior data; Inputting the user behavior data into an active analysis model to perform user profiling and obtain the behavior preference information of the target user; the active analysis model is used to mine and analyze the behavior preference characteristics in the user behavior data from multiple dimensions; Based on the behavior preference information and the current geographic location of the target user, predicting the target user's behavior trajectory information within a preset future time period through a behavior construction model; the behavior trajectory information includes behavior nodes of the target user whose predicted execution probability reaches a preset threshold within the preset future time period; Based on the behavior trajectory information, a comprehensive service unit matching the current geographic location is recommended to the target user; the comprehensive service unit includes a service unit matching the behavior node in the behavior trajectory information, and the service unit is at least one of an offline product, an offline store, an offline service business, an online service business, and an online product; The active analysis model includes at least the following structures: an extraction layer, a construction layer, an analysis layer, a prediction layer, and an output layer; inputting the user behavior data into the active analysis model to perform user profiling and obtain the behavior preference information of the target user includes: Extracting multi-dimensional user behavior features from the user behavior data through an extraction layer; By constructing layers, the multidimensional user behavior features are projected into a multidimensional portrait space according to the projection strategies corresponding to the respective dimensions to obtain a corresponding multidimensional user portrait; wherein each portrait coordinate point in the multidimensional user portrait corresponds to at least one behavior feature point in the multidimensional user behavior features; The multidimensional user profile is adaptively segmented using pre-configured slice fitting parameters through the analysis layer to obtain multiple behavioral slices corresponding to the multidimensional user profile; wherein the multiple behavioral slices are respectively used to represent the behavioral change characteristics of the multidimensional user profile under different analysis angles; the slice fitting parameters are used to indicate the slicing method of the multidimensional user profile and the method of selecting the slicing positions; Through the prediction layer, the matching probability between multiple behavior slices and preset behavior types is predicted to select candidate behavior types for the multi-dimensional user portrait; Through the output layer, the behavior preference information of the target user is generated according to the candidate behavior type.

2. The large-scale model recommendation method based on individual user behavior according to claim 1 is characterized in that: The prediction layer predicts the matching probability between multiple behavior slices and preset behavior types to select candidate behavior types for the multi-dimensional user portrait, including: For the jth preset behavior type among multiple preset behavior types , calculate multiple behavior slices and preset behavior types The matching degree between ; Among them, the i-th behavior slice With preset behavior type between It is expressed as the following formula: , Slice for behavior With preset behavior type The cosine similarity of Slice for behavior With preset behavior type The matching probability vector obtained by the preset neural network processing, is the number of behavior slices, Slice for behavior The corresponding weight coefficient; Combine multiple preset behavior types with preset behavior types The matching degree between them is summarized to obtain multiple preset behavior types and preset behavior types. The matching probability between ; A candidate behavior type whose matching probability meets the preset screening condition is selected from multiple preset behavior types.

3. The large-scale model recommendation method based on individual user behavior according to claim 2 is characterized in that: Multiple preset behavior types and preset behavior types The matching probability between It is expressed as the following formula: in, and Behavior Slices and another row for the slice The mean of is an exponential term based on the covariance matrix, which is used to represent the behavior slice and another row for the slice The correlation between them.

4. The large-scale model recommendation method based on individual user behavior according to claim 1 is characterized in that: The analysis layer uses pre-configured slice fitting parameters to adaptively segment the multi-dimensional user profile to obtain multiple behavior slices corresponding to the multi-dimensional user profile, including: Based on the selection method of the slice interception position, set the starting position and the ending position of each behavior slice; Obtain the computational complexity corresponding to each behavior slice under different analysis angle combinations, and select the target analysis angle combination that matches each behavior slice based on the obtained computational complexity; Based on the starting position and ending position of each behavior slice, the corresponding slice perspectives are combined from the perspective of target analysis to adaptively segment the multi-dimensional user portrait to obtain multiple behavior slices; wherein the adaptive segmentation is at least one of equidistant segmentation, peak segmentation based on data distribution, and cluster segmentation.

5. The large-scale model recommendation method based on individual user behavior according to claim 1 is characterized in that: The method of predicting the target user's behavior trajectory information within a preset time period in the future based on the behavior preference information and the target user's current geographical location through a behavior building model includes: Get the target user's current geographic location; Based on the current geographic location and the mapping relationship between the geographic location type and the duration configuration strategy, set the future preset duration corresponding to the target user; Obtain corresponding surrounding environment data based on the current geographical location; Based on the configured future preset duration, the behavior preference information, the surrounding environment data, and the current geographical location are input into the behavior construction model to perform behavior trajectory prediction to obtain the behavior trajectory information.

6. The large-scale model recommendation method based on individual user behavior according to claim 5 is characterized in that: The behavior construction model includes at least the following structures: extraction layer, fusion layer, prediction layer, and screening layer; The future preset time duration obtained based on the configuration is inputted into the behavior construction model to predict the behavior trajectory, and the behavior trajectory information is obtained, including: extracting behavior preference features from the behavior preference information and extracting surrounding environment features from the surrounding environment data through an extraction layer; The behavior preference characteristics, the surrounding environment characteristics, and the current geographical location are associated and fused through a fusion layer to obtain a fused behavior characteristic of the target user; the fused behavior characteristic includes associations of multiple combinations of each behavior preference characteristic with different surrounding environment characteristics and different geographical locations; Through the prediction layer, a multi-branch decision layer is used to predict the changes of the fused behavior features in the future preset time length, and obtain the predicted behavior nodes of the target user in the future preset time length and the corresponding predicted execution probability; Through the screening layer, the predicted behavior nodes with the preset threshold of the predicted execution probability are constructed as the behavior trajectory information.

7. A large-scale model recommendation system based on individual user behavior, characterized by: The system comprises at least the following units: An acquisition unit, used to acquire user behavior data of a target user; The user behavior data includes at least: online behavior data and offline behavior data; An analysis unit is configured to input the user behavior data into an active analysis model to perform user profiling and obtain the behavior preference information of the target user; the active analysis model is configured to mine and analyze the behavior preference features in the user behavior data from multiple dimensions; A prediction unit is configured to predict the target user's behavior trajectory information within a preset future time period based on the behavior preference information and the target user's current geographic location through a behavior construction model; the behavior trajectory information includes behavior nodes of which the target user's predicted execution probability reaches a preset threshold within the preset future time period; a recommendation unit, configured to recommend, to the target user, a comprehensive service unit that matches the current geographic location based on the behavior trajectory information; the comprehensive service unit includes a service unit that matches the behavior node in the behavior trajectory information, and the service unit is at least one of an offline product, an offline store, an offline service business, an online service business, and an online product; The active analysis model includes at least the following structures: an extraction layer, a construction layer, an analysis layer, a prediction layer, and an output layer. The analysis unit inputs the user behavior data into the active analysis model to perform user profiling and obtain the behavior preference information of the target user, specifically for: Extracting multi-dimensional user behavior features from the user behavior data through an extraction layer; By constructing layers, the multidimensional user behavior features are projected into a multidimensional portrait space according to the projection strategies corresponding to the respective dimensions to obtain a corresponding multidimensional user portrait; wherein each portrait coordinate point in the multidimensional user portrait corresponds to at least one behavior feature point in the multidimensional user behavior features; The multidimensional user profile is adaptively segmented using pre-configured slice fitting parameters through the analysis layer to obtain multiple behavioral slices corresponding to the multidimensional user profile; wherein the multiple behavioral slices are respectively used to represent the behavioral change characteristics of the multidimensional user profile under different analysis angles; the slice fitting parameters are used to indicate the slicing method of the multidimensional user profile and the method of selecting the slicing positions; Through the prediction layer, the matching probability between multiple behavior slices and preset behavior types is predicted to select candidate behavior types for the multi-dimensional user portrait; Through the output layer, the behavior preference information of the target user is generated according to the candidate behavior type.

8. An electronic device, characterized in that: including a memory for storing a computer software program; A processor is used to read and execute the computer software program, thereby implementing the various steps in the large model recommendation method based on personal user terminal behavior as described in any one of claims 1-6.

9. A non-transitory computer-readable storage medium, characterized in that The storage medium stores a computer software program, which, when executed by a processor, implements the various steps of the large-scale model recommendation method based on individual user-side behavior as described in any one of claims 1 to 6.

Images