Community group identification method and device, equipment, medium and product

Abstract

The invention discloses a community group identification method and device, equipment, a medium and a product, and relates to the technical field of data processing. The community group identification method comprises the steps of obtaining user portrait features of users in a community, time-space association data and target interest points in the community, wherein the target interest points are used for representing interest points corresponding to a group type of a target community group; based on the user portrait features and the time-space association data, performing density clustering on the users in the community to obtain a first group identification result corresponding to the group type; matching the space-time association data of each user in the community with the position of the target interest point to obtain a second group identification result corresponding to the group type; and fusing the first group identification result and the second group identification result to obtain a target group identification result corresponding to the group type in the community.

Description

Technical Field

[0001] The present application belongs to the field of data processing technology, and in particular, relates to a method, device, equipment, medium and product for identifying a community group. Background Art

[0002] In a smart community, people's daily life, work or other activities will form community groups, and the aggregation analysis of community groups through time and space correlation plays an extremely important role in the management of smart communities and even the governance of smart cities. Therefore, it is very important to accurately identify community groups in the community area.

[0003] In the existing method, outdoor camera equipment is used to directly capture the on-site images of the target scene. The on-site images are transmitted back to the background, which identifies and judges the situation of the people, thereby determining the situation of the community group.

[0004] However, this method only identifies community groups through on-site images of the target scene, and the factors considered in the identification process are relatively simple, and the identification accuracy of community groups is not high. Summary of the invention

[0005] The embodiments of the present application provide a method, device, equipment, medium and product for identifying a community group, which can improve the accuracy of identifying a community group.

[0006] In one aspect of an embodiment of the present application, a method for identifying a community group is provided, the method comprising:

[0007] Obtain user portrait features, spatiotemporal correlation data, and target interest points in the community for users in the community, where the target interest points are used to represent each interest point corresponding to the group type of the target community group;

[0008] Based on user portrait features and spatiotemporal correlation data, density clustering is performed between users in the community to obtain a first group identification result corresponding to the group type;

[0009] Matching the spatiotemporal correlation data of each user in the community with the location of the target point of interest to obtain a second group recognition result corresponding to the group type;

[0010] The first group identification result is fused with the second group identification result to obtain a target group identification result corresponding to the group type in the community.

[0011] In one aspect of an embodiment of the present application, a device for identifying a community group is provided, the device comprising:

[0012] An information acquisition module is used to obtain user portrait features, spatiotemporal correlation data, and target interest points in the community, where the target interest points are used to represent each interest point corresponding to the group type of the target community group;

[0013] A user clustering module is used to perform density clustering between users in the community based on user portrait features and spatiotemporal correlation data to obtain a first group identification result corresponding to the group type;

[0014] A user matching module is used to match the spatiotemporal correlation data of each user in the community with the location of the target interest point to obtain a second group recognition result corresponding to the group type;

[0015] The result fusion module is used to fuse the first group identification result with the second group identification result to obtain a target group identification result corresponding to the group type in the community.

[0016] According to one aspect of an embodiment of the present application, an electronic device is provided, which includes: a memory and a program or instruction stored in the memory and executable on a processor, wherein when the program or instruction is executed by the processor, a method for identifying a community group as provided in any aspect of the above-mentioned embodiment of the present application is implemented.

[0017] In one aspect of an embodiment of the present application, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, a method for identifying a community group as provided in any aspect of the above-mentioned embodiment of the present application is implemented.

[0018] According to one aspect of an embodiment of the present application, a computer program product is provided. When instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes a method for identifying a community group as provided in any aspect of the above-mentioned embodiment of the present application.

[0019] In the community group identification method provided in the embodiment of the present application, the user portrait features, spatiotemporal correlation data and target points of interest in the target community of each user in the community are comprehensively obtained. Starting from the individual user, a first group identification result corresponding to the group type is determined according to the user portrait features and spatiotemporal correlation data; starting from the overall scene in the community, a second group identification result corresponding to the group type is determined according to the spatiotemporal correlation data of each user in the community and the location of the target point of interest in the community. Finally, the first group identification result is fused with the second group identification result to obtain the final target group identification result. In this way, the present application can improve the recognition accuracy of the community group by starting from the two dimensions of individual users and the overall scene in the community and fusing the final target group identification result. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solution of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0021] Figure 1 It is a flowchart of a method for identifying a community group provided by an embodiment of the present application;

[0022] Figure 2 It is a schematic diagram of the structure of a community group identification device provided by an embodiment of the present application;

[0023] Figure 3 It is a schematic diagram of the structure of a community group identification device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0024] The features and exemplary embodiments of each aspect of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.

[0025] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0026] It should be noted that the acquisition, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of national laws and regulations.

[0027] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0028] In the existing method, outdoor cameras are used to directly capture live images of the target scene, and the live images are sent back to the background, which identifies and judges the situation of the people, thereby determining the situation of the community group. However, this method only identifies the community group through the live images of the target scene, and the factors considered in the identification process are relatively single, and the recognition accuracy of the community group is not high.

[0029] The purpose of the present application is to provide a method, device, equipment, medium and product for identifying community groups. In the method for identifying community groups provided in the embodiment of the present application, the user portrait features, spatiotemporal correlation data and target points of interest in the target community of each user in the community are comprehensively obtained. Starting from the individual user, a first group identification result corresponding to the group type is determined according to the user portrait features and spatiotemporal correlation data; starting from the overall scene in the community, a second group identification result corresponding to the group type is determined according to the spatiotemporal correlation data of each user in the community and the location of the target point of interest in the community. Finally, the first group identification result is fused with the second group identification result to obtain the final target group identification result. In this way, the present application can improve the recognition accuracy of the community group by starting from the two dimensions of individual users and the overall scene in the community and fusing the final target group identification result.

[0030] The following describes specific embodiments of the community group identification method, device, equipment, medium and product provided in the embodiments of the present application. The following describes specific embodiments of the community group identification method provided in the embodiments of the present application.

[0031] like Figure 1 As shown, a flow chart of a method for identifying a community group is provided. The method for identifying a community group can be applied to a server, and the method for identifying a community group can include the following S101 to S104.

[0032] S101, obtaining user portrait features, spatiotemporal correlation data and target interest points in the community, where the target interest points are used to represent each interest point corresponding to a group type of a target community group.

[0033] In this embodiment, the user portrait feature is used to characterize a series of attributes and tags that provide a comprehensive, detailed, and in-depth description of the user. The user portrait feature is intended to reveal the user's preferences, behavior patterns, needs, and other relevant information, so as to more accurately determine the community group to which the user belongs. Specifically, the user portrait feature can be derived from the user's registration information, online behavior records, and social media activities.

[0034] For example, user portrait features may include: user number, gender, age, terminal type, number of community access points, application usage preferences, web browsing preferences, social interaction preferences, community application usage preferences, community recommendation acceptance, number of roaming times, number of service subscriptions, content consumption preferences, family / enterprise user identification, loyalty program participation, user network age, user account opening location, user basic package type 4G / 5G, user package name code, basic package traffic, whether ordinary traffic package is activated, traffic included in ordinary traffic package, off-peak traffic, total traffic, billing fee / yuan, out-of-package traffic income / yuan, mobile TV traffic, WeChat traffic usage, the cell with the largest user traffic generation in the month, the cell with the largest number of user call minutes in the month, the cell with the most user calls in the month, traffic generated on 4G network, number of call minutes generated on 4G network, traffic generated on 5G network, number of call minutes generated on 5G network , participation in the contract plan, contract expiration date, whether it is a joint account user, whether it is a member of the integrated service, number of users contacted by calls, user level, total duration of calling and called parties_minutes, total number of calls in the month, number of calls on different networks, number of days of traffic use per month, available traffic in the user's main package, the maximum number of calls of the user in all communities in the month, whether it is a usim card, gender, whether the 5G Internet access function is activated, user star rating, number of traffic Internet access services on the 5G network, number of voice call services on the 5G network, number of traffic Internet access services on the 4G network, number of voice call services on the 4G network, whether the 4G copy package function package is activated, total duration of calls / minutes, income from voice minutes exceeding the package / yuan, number of voice minutes included in the user's main package, free traffic included in the user package, number of traffic fast food packages activated in the month, total number of voice minutes used in the month, number of text messages sent in the month, number of text messages received in the month, etc.

[0035] The spatiotemporal correlation data is used to characterize the time and space information of the user. The time information includes the time point field of the sampling point, and the space information includes the longitude information, latitude information and altitude information of the user. For example, the spatiotemporal correlation data may include Minimization Drive Test (MDT) data or Over The Top (OTT) data.

[0036] Specifically, the server can extract the latitude and longitude information and altitude information of the user's location through data analysis in the communication interface between the base station and the packet core network, and associate it with the mobile communication data at this moment according to the user identification, thereby obtaining the user's spatiotemporal correlation data.

[0037] The target points of interest are the points of interest in the community that correspond to the group type of the target community group. The target community group is used to characterize the community group that meets the identification requirements of the community group to be constructed. As an example, the target community group can be a high-end dining community group, and the target points of interest at this time are the high-end dining places in the community.

[0038] Specifically, the server collects points of interest in the community in advance and organizes each point of interest according to a certain classification system. For example, it can be divided into multiple categories such as catering, shopping, accommodation, travel, sports and entertainment, and financial services. Each category can be further divided into multiple subcategories. Then, after determining the group type of the target community group, the server obtains the corresponding target points of interest from the pre-collected points of interest.

[0039] S102: Based on the user portrait features and the spatiotemporal correlation data, density clustering is performed on the users in the community to obtain a first group identification result corresponding to the group type.

[0040] In this embodiment, the first group identification result is used to represent the group identification result corresponding to the group type of the target community group determined by analyzing the individual user according to the user portrait features and spatiotemporal correlation data.

[0041] As an example, the server extracts key features related to the group type of the target community group from user portrait features and spatiotemporal correlation data. These features can reflect the similarities and differences between users. Then, a density clustering algorithm (such as the DBSCAN algorithm, the HDBSCAN algorithm, etc.) is used to cluster the users in the community. These algorithms can identify the close relationships between users and classify similar users into the same group. Finally, based on the clustering results, a first group identification result corresponding to the group type is obtained.

[0042] S103, matching the spatiotemporal correlation data of each user in the community with the location of the target point of interest to obtain a second group recognition result corresponding to the group type.

[0043] In this embodiment, the second group identification result is used to represent the group identification result corresponding to the group type of the target community group determined based on the overall scene in the community and the spatiotemporal correlation data of each user in the community and the location of the target interest point in the community.

[0044] As an example, the server matches the user's spatiotemporal association data (such as location trajectory) with the location of the target point of interest. Then, based on the user's behavioral characteristics when visiting the target point of interest (such as visit frequency, length of stay, etc.), the user's preference and interest in the target point of interest is analyzed. Based on the association analysis results between the user and the target point of interest, a second group identification result corresponding to the group type of the target community group is obtained.

[0045] S104: The first group identification result and the second group identification result are integrated to obtain a target group identification result corresponding to the group type in the community.

[0046] In this embodiment, the target group identification result is used to characterize the group identification result corresponding to the group type of the target community group that is finally obtained.

[0047] As an example, the server merges the first group identification result with the second group identification result. Specifically, a weighted average fusion strategy may be adopted. After fusion, a target group identification result corresponding to the group type of the target community group in the community is obtained.

[0048] As another example, when the group identification results include group user information, group high-frequency time periods, and group high-frequency areas, the group user information, group high-frequency time periods, and group high-frequency areas in the first group identification results are respectively unioned with the group user information, group high-frequency time periods, and group high-frequency areas in the second group identification results to obtain a target group identification result corresponding to the group type of the target community group in the community.

[0049] In the community group identification method provided in this embodiment, the user portrait features, spatiotemporal correlation data, and target interest points in the target community of each user in the community are comprehensively obtained. Starting from the individual user, a first group identification result corresponding to the group type is determined based on the user portrait features and spatiotemporal correlation data; starting from the overall scene in the community, a second group identification result corresponding to the group type is determined based on the spatiotemporal correlation data of each user in the community and the location of the target interest point in the community. Finally, the first group identification result is fused with the second group identification result to obtain the final target group identification result. In this way, the present application can improve the recognition accuracy of community groups by starting from the two dimensions of individual users and the overall scene in the community to obtain the final target group identification result.

[0050] As an optional embodiment, S102 may specifically include:

[0051] Inputting the user portrait features of each user into a pre-trained user recognition model to obtain a first user corresponding to the group type;

[0052] Splitting the spatiotemporal correlation data of the first user into a plurality of first spatiotemporal correlation data slices according to a preset time interval;

[0053] Density clustering is performed on each first spatiotemporal correlation data slice in the same time interval to obtain a spatiotemporal cluster in each time interval, where the time length of the time interval is equal to a preset time interval;

[0054] Statistical analysis is performed on the spatiotemporal clusters in each time interval to obtain a first group recognition result corresponding to the group type.

[0055] In this embodiment, the user identification model is used to identify whether the user belongs to the group type corresponding to the target community group. The first user is used to characterize the user who belongs to the group type corresponding to the target community group.

[0056] As an example, the server first trains a user identification model based on the historical data of users in the target community group. The user identification model can be a classifier based on machine learning or deep learning, such as support vector machine (SVM), random forest, neural network, etc., which aims to distinguish whether a user belongs to the target community group based on user portrait features.

[0057] Then, the newly collected user portrait features are input into the pre-trained user identification model. The user identification model outputs the probability that the corresponding user belongs to the target community group. When the probability of belonging to the target community group is greater than the preset probability threshold p, the user is determined to be the first user belonging to the target community group.

[0058] Then, for the set of identified first users, the spatiotemporal correlation data of each first user is obtained. Specifically, it includes the geographic location information (such as GPS coordinates) of the first user at different time points, the types of places visited (such as shopping malls, restaurants, parks), and the activity records (such as consumption, stay time) at these places. And according to the preset time intervals (such as every day, every hour, every minute), the spatiotemporal correlation data is split into multiple first spatiotemporal correlation data slices. Each first spatiotemporal correlation data slice represents the spatiotemporal behavior of the user in a specific time period.

[0059] Then, multiple time intervals are determined according to the time length corresponding to the preset time intervals. A density clustering algorithm is applied to each first spatiotemporal correlation data slice in each time interval to obtain a spatiotemporal cluster in each time interval. Each spatiotemporal cluster represents a user group with similar spatiotemporal behavior patterns in the same time period.

[0060] Finally, statistical analysis is performed on the spatiotemporal clusters in each time interval, including calculating the size of the spatiotemporal clusters (i.e., the number of users included), the diversity of the visited locations, etc. Based on the results of the statistical analysis, a first group identification result corresponding to the group type of the target community group is obtained. Specifically, it can include identifying the active time period of the target community group, the main activity area of ​​the target community group, and the common characteristics or behavior patterns of the target community group.

[0061] Through this embodiment, based on user portrait features and spatiotemporal correlation data, density clustering is performed on users in the community to obtain a first group identification result corresponding to the group type. In this way, starting from individual users, mining and analysis based on user portrait features and spatiotemporal correlation data can improve the accuracy of community group identification.

[0062] As an optional embodiment, before inputting the user portrait features of each user into a pre-trained user recognition model to obtain the first user corresponding to the group type, the community group recognition method may further include:

[0063] Obtain a training sample set, the training sample set includes multiple training samples, each training sample includes a user portrait feature and a corresponding user type label, the user type label includes a user label belonging to a group type and a user label not belonging to a group type;

[0064] The target model is trained using the training samples until the training stop condition is met to obtain a user recognition model.

[0065] In this embodiment, the training sample set is composed of multiple training samples, each of which contains user portrait features and their corresponding user type labels. The user portrait features can be obtained through data mining, questionnaire surveys, user behavior tracking, etc., and the user type labels can be generated through expert annotation, machine learning algorithms, etc.

[0066] Then, the server selects a target model suitable for user identification, such as a deep learning model (such as CNN, ResNet, etc.) or other machine learning models, and divides the training sample set into a training set and a validation set. The target model is trained with the training set, and the target model is validated with the validation set. The training error is minimized by adjusting the parameters of the target model until the training stop condition is reached. The parameters and structure of the current target model are saved to obtain a user identification model.

[0067] Among them, the preset training stop conditions may be reaching a preset number of training rounds or iterations, the recognition probability on the validation set no longer decreasing or starting to increase, the performance of the target model on the validation set reaching a preset accuracy rate, etc.

[0068] Through this embodiment, a training sample set corresponding to the group type of the target community group is obtained, and the training sample is used to train the target model to obtain a user identification model. In this way, through the user identification model, it is possible to accurately determine whether each user belongs to the first user corresponding to the group type of the target community group, which can improve the recognition accuracy of the first user.

[0069] As an optional embodiment, density clustering is performed on each first spatiotemporal correlation data slice in the same time interval to obtain a spatiotemporal cluster in each time interval, which may specifically include:

[0070] The model parameters of the initial clustering model are set to obtain the target clustering model;

[0071] Calculating the interval distances between the position points corresponding to the first spatiotemporal correlation data slices according to the first spatiotemporal correlation data slices;

[0072] According to each interval distance, density clustering is performed on each first spatiotemporal correlation data slice in the same time interval through a target clustering model to obtain a spatiotemporal cluster in each time interval.

[0073] In this embodiment, the server first selects a suitable clustering model as the initial clustering model, which needs to support density clustering (such as DBSCAN, DENCLUE, etc.). And according to actual needs and data characteristics, the parameters of the initial clustering model are set. The initial clustering model with set parameters is used as the target clustering model for subsequent density clustering.

[0074] For example, for the DBSCAN model, parameters such as neighborhood radius, minimum number of points, and measurement method need to be set. Among them, the neighborhood radius is used to determine the neighborhood range of a point, which indicates the maximum distance between two sample points considered as adjacent; the minimum number of points refers to the minimum number of neighboring points required for a point to be considered as a core point within the neighborhood radius. Only when the number of points in a point's neighborhood is greater than or equal to the minimum number of points, the point is considered a core point, so that clustering can be expanded; the measurement method is a measurement method used to calculate distance, such as Euclidean distance, Manhattan distance, etc.

[0075] Then, according to each first spatiotemporal correlation data slice, a corresponding position point is extracted from each first spatiotemporal correlation data slice, and the interval distance between the position points corresponding to each first spatiotemporal correlation data slice is calculated.

[0076] Finally, for each first spatiotemporal correlation data slice in the same time interval, density clustering is performed using the target clustering model. In the density clustering process, spatiotemporal clusters in each time interval are gradually identified and extracted.

[0077] Through this embodiment, a target clustering model is set to cluster according to the interval distance between the location points corresponding to each first spatiotemporal association data slice, and obtain spatiotemporal clusters in each time interval. This helps to perform statistical analysis on the spatiotemporal clusters in each time interval and obtain a first group identification result corresponding to the group type of the target community group.

[0078] As an optional embodiment, the first spatiotemporal correlation data slice includes location point coordinates, and the location point coordinates include longitude coordinates, latitude coordinates, and altitude coordinates;

[0079] Calculating the interval distances between the position points corresponding to the first spatiotemporal correlation data slices according to the first spatiotemporal correlation data slices may specifically include:

[0080] Determine a horizontal distance between the first location point and the second location point according to the first longitude coordinate and the first latitude coordinate of the first location point, and the second longitude coordinate and the second latitude coordinate of the second location point, wherein the first location point and the second location point are location points corresponding to any two different first spatiotemporal correlation data slices;

[0081] The first altitude coordinate of the first location point is subtracted from the second altitude coordinate of the second location point and the absolute value is taken to obtain the altitude distance between the first location point and the second location point;

[0082] The interval distance between the first location point and the second location point is determined according to the horizontal distance and the altitude distance.

[0083] In this embodiment, the server obtains the first longitude coordinate, the first latitude coordinate and the first altitude coordinate of the first location point, and the second longitude coordinate, the second latitude coordinate and the second altitude coordinate of the second location point from the first spatiotemporal correlation data slice.

[0084] Then, the longitude difference between the first location point and the second location point and the latitude difference between the first location point and the second location point are calculated. According to the longitude difference and the latitude difference, the horizontal distance between the first location point and the second location point is calculated using the spherical distance formula.

[0085] Then, the altitude difference between the first location point and the second location point is calculated, and the absolute value of the altitude difference is taken to obtain the altitude distance between the first location point and the second location point.

[0086] Finally, the square value of the horizontal distance between the first location point and the second location point, and the square value of the altitude distance between the first location point and the second location point are calculated, and the square value of the horizontal distance is added to the square value of the altitude distance and then squared to obtain the interval distance between the first location point and the second location point.

[0087] Through this embodiment, the interval distance between the first location point and the second location point is calculated based on the first longitude coordinate, the first latitude coordinate and the first altitude coordinate of the first location point, and the second longitude coordinate, the second latitude coordinate and the second altitude coordinate of the second location point. In this way, the interval distance between the first location point and the second location point can be accurately calculated, which helps to achieve accurate density clustering of the first spatiotemporal correlation data slice according to the interval distance, thereby improving the accuracy of density clustering.

[0088] As an optional embodiment, determining the horizontal distance between the first location point and the second location point according to the first longitude coordinate and the first latitude coordinate of the first location point, and the second longitude coordinate and the second latitude coordinate of the second location point may specifically include:

[0089] The target cosine value, the cosine value of the first latitude coordinate, and the cosine value of the second latitude coordinate are multiplied to obtain a cosine calculation result, where the target cosine value is the cosine value of the difference between the first longitude coordinate and the second longitude coordinate;

[0090] Multiply the sine value of the first latitude coordinate and the sine value of the second latitude coordinate to obtain a sine calculation result;

[0091] The accumulated value of the cosine calculation result and the sine calculation result is arc-cosine-calculated and then multiplied by the target radius to obtain the horizontal distance between the first position point and the second position point.

[0092] In this embodiment, the target radius is used to represent the radius of the earth. Specifically, the target radius is equal to 6371*1000 meters.

[0093] The horizontal distance between the first position point and the second position point can be specifically determined by the following formula 1:

[0094] d s =R*arcos[cos(y1)*cos(y2)*cos(x1-x2)+sin(y1)*sin(y2)] Formula 1

[0095] Where, d s Used to represent the horizontal distance between the first location point and the second location point, R is used to represent the radius of the earth, arcos is used to represent the inverse cosine calculation, cos(y1) is used to represent the cosine value of the first latitude coordinate, cos(y2) is used to represent the cosine value of the second latitude coordinate, cos(x1-x2) is used to represent the cosine value of the difference between the first longitude coordinate and the second longitude coordinate, sin(y1) is used to represent the sine value of the first latitude coordinate, and sin(y2) is used to represent the sine value of the second latitude coordinate.

[0096] Through this embodiment, the horizontal distance between the first location point and the second location point is accurately calculated based on the first longitude coordinate and the first latitude coordinate of the first location point, and the second longitude coordinate and the second latitude coordinate of the second location point. This helps to accurately calculate the interval distance between the first location point and the second location point based on the horizontal distance, thereby achieving accurate density clustering of the first spatiotemporal correlation data slice and improving the accuracy of density clustering.

[0097] As an optional embodiment, density clustering is performed on each first spatiotemporal correlation data slice in the same time interval by using a target clustering model according to each interval distance to obtain a spatiotemporal cluster cluster in each time interval, which may specifically include:

[0098] Constructing a core point set according to the interval distances between the position points corresponding to the first spatiotemporal correlation data slices in the same time interval, wherein the core point set includes at least one core point among the position points corresponding to the first spatiotemporal correlation data slices;

[0099] The following steps are executed repeatedly until each core point in the core point set is traversed to obtain multiple spatiotemporal clusters in the time interval:

[0100] Select a target core point from the core point set, where the target core point is any core point that has not been selected;

[0101] The target core point is stored in the initial cluster, and the directly density reachable points corresponding to the target core point are stored in an ordered queue. The directly density reachable points in the ordered queue are sorted in ascending order according to the reachable distance between them and the target core point.

[0102] Obtain directly density-reachable points from the ordered queue in order, and determine whether the directly density-reachable points belong to core points;

[0103] When the directly density reachable point belongs to the core point, the target core point is updated to a directly density reachable point, and the loop execution is returned to store the target core point in the spatiotemporal cluster, and the directly density reachable points corresponding to the target core point are stored in the ordered queue until all the directly density reachable points in the ordered queue are traversed to obtain the spatiotemporal cluster in the time interval.

[0104] In this embodiment, when the number of points contained in the neighborhood radius of a point is greater than or equal to the minimum number of points, the point is a core point.

[0105] Direct density reachable points are used to represent points within the neighborhood radius corresponding to the core point.

[0106] The reachable distance is used to characterize the maximum value between the core distance of the core point corresponding to the directly density reachable point and the Euclidean distance between the directly density reachable point and the core point. The core distance is used to characterize the minimum neighborhood radius that makes the target point become a core point.

[0107] As an example, the server selects core points among the position points corresponding to each first spatiotemporal correlation data slice according to the interval distance between the position points corresponding to each first spatiotemporal correlation data slice in the same time interval to form a core point set.

[0108] Then, create an initial clustering cluster for storing the clustering cluster currently being processed, and an ordered queue. From the core point set, select a core point that has not been selected as the target core point. Add the target core point to the initial clustering cluster, and calculate all the direct density reachable points of the core point. These direct density reachable points are sorted from small to large according to the reachable distance between them and the target core point, and put into the ordered queue.

[0109] Then, starting from the head of the ordered queue, take out the directly density-reachable points in turn. Check whether the point is a core point: if it is a core point, add the point to the initial clustering cluster, and regard the point as a new target core point, find all the directly density-reachable points corresponding to the point, put these points into the ordered queue, and reorder the points in the ordered queue according to the reachable distance; if it is not a core point, continue to get the next point from the ordered queue.

[0110] When all points in the ordered queue are processed, the currently processed initial cluster is a complete spatiotemporal cluster. Then, the loop is executed to select a core point that has not been selected from the core point set as the target core point until all core points in the core point set are traversed.

[0111] Through this embodiment, according to each interval distance, each first spatiotemporal correlation data slice in the same time interval is density clustered by the target clustering model to obtain a spatiotemporal cluster cluster in each time interval. This helps to perform statistical analysis on the spatiotemporal cluster clusters later, thereby accurately determining the first group recognition result corresponding to the group type.

[0112] As an optional embodiment, the first group identification result includes first group user information, first group high frequency time period and first group high frequency area;

[0113] Performing statistical analysis on the spatiotemporal clusters in each time interval to obtain a first group recognition result corresponding to the group type may specifically include:

[0114] Traversing the spatiotemporal clusters in each time interval, and determining the spatiotemporal clusters whose occurrence frequency is greater than or equal to a preset frequency threshold as frequent clusters;

[0115] Performing a union operation on each frequent cluster to obtain first group user information corresponding to the group type;

[0116] According to the time interval corresponding to each frequent cluster, the high-frequency time period of the first group corresponding to the group type is obtained by statistics;

[0117] Each frequent cluster is projected into the community after three-dimensional rasterization, and the high-frequency area of ​​the first group corresponding to the group type is obtained by statistics.

[0118] In this embodiment, the server sets a frequency threshold according to actual needs to filter frequently occurring spatiotemporal clusters, and then compares the occurrence frequency of each spatiotemporal cluster with the preset frequency threshold, and determines the clusters with a frequency greater than or equal to the threshold as frequent clusters.

[0119] Then, a union operation is performed on each frequent cluster, and overlapping clusters are merged to obtain a comprehensive user information set representing a specific group type, namely, the first group user information; the first a time periods with the highest frequency of spatiotemporal clusters are counted with a time length greater than that corresponding to the time interval as the time granularity to obtain the first group high-frequency time period related to the group type.

[0120] Finally, each frequent cluster is projected into the community model after three-dimensional rasterization, and the top b grids with the highest frequency of spatiotemporal clusters are counted to obtain the high-frequency area of ​​the first group related to the group type. Among them, three-dimensional rasterization is to divide the community space into three-dimensional grids, each grid represents a certain spatial range and height (such as floor), so as to facilitate a more detailed analysis of user activity areas.

[0121] Through this embodiment, the spatiotemporal clusters in each time interval are statistically analyzed to obtain the first group user information, the first group high-frequency time period, and the first group high-frequency area corresponding to the group type of the target community group. In this way, the present application can improve the recognition accuracy of the community group by analyzing from three aspects: user information, high-frequency time period, and high-frequency area.

[0122] As an optional embodiment, S103 may specifically include:

[0123] Splitting the spatiotemporal correlation data of each user in the community into a plurality of second spatiotemporal correlation data slices according to a preset time interval;

[0124] Matching the location points corresponding to each second spatiotemporal correlation data slice with the location of the target point of interest, and obtaining the number of occurrences and the frequency of occurrence of each user in the community at the target point of interest;

[0125] When the number of occurrences and the frequency of occurrences meet a preset statistical condition, determining that the user belongs to a second user corresponding to the group type;

[0126] Statistical analysis is performed on the spatiotemporal correlation data of each second user to obtain a second group recognition result corresponding to the group type.

[0127] In this embodiment, the server splits the spatiotemporal correlation data into multiple second spatiotemporal correlation data slices according to preset time intervals (such as every day, every hour, every minute). Each second spatiotemporal correlation data slice represents the spatiotemporal behavior of a user in a specific time period.

[0128] Then, the location points corresponding to each second spatiotemporal correlation data slice are matched with the location of the target interest point to determine whether the user appears in the area corresponding to the target interest point. The matching results are counted to obtain the number of occurrences and the frequency of occurrence of each user in the community at the target interest point. According to the preset statistical conditions (such as the number of occurrences threshold, the frequency of occurrence threshold, etc.), the users who meet the conditions are screened out and determined as the second users corresponding to the group type.

[0129] Finally, statistical analysis is performed on the second spatiotemporal correlation data slices of each second user to obtain the second group identification result corresponding to the group type. Specifically, statistics are performed on each second user to obtain the second group user information; the first a time periods with the highest frequency of the second spatiotemporal correlation data slices are counted with a time length greater than the time interval corresponding to the time granularity to obtain the second group high-frequency time period related to the group type; each second spatiotemporal correlation data slice is projected into the community model after three-dimensional rasterization, and the first b grids with the highest frequency of the second spatiotemporal correlation data slices are counted to obtain the second group high-frequency area related to the group type.

[0130] Through this embodiment, the second group recognition result corresponding to the group type is determined based on the spatiotemporal correlation data of each user in the community and the location of the target interest point in the community. In this way, starting from the overall scene in the community, mining and analysis based on the spatiotemporal correlation data and the location of the target interest point in the community can improve the recognition accuracy of the community group.

[0131] Community-based identification method. Accordingly, the present application also provides a specific embodiment of a community-based identification device.

[0132] like Figure 2 As shown, the community group identification device 200 provided in the embodiment of the present application includes an information acquisition module 210 , a user clustering module 220 , a user matching module 230 and a result fusion module 240 .

[0133] The information acquisition module 210 is used to obtain user portrait features, spatiotemporal correlation data and target interest points in the community of users in the community. The target interest points are used to represent each interest point corresponding to the group type of the target community group.

[0134] The user clustering module 220 is used to perform density clustering on users in the community based on user portrait features and spatiotemporal correlation data to obtain a first group identification result corresponding to the group type.

[0135] The user matching module 230 is used to match the spatiotemporal correlation data of each user in the community with the location of the target interest point to obtain a second group recognition result corresponding to the group type.

[0136] The result fusion module 240 is used to fuse the first group identification result with the second group identification result to obtain a target group identification result corresponding to the group type in the community.

[0137] As an optional embodiment, the user clustering module 220 specifically includes the following units:

[0138] A feature input unit, used to input the user portrait feature of each user into a pre-trained user recognition model to obtain a first user corresponding to the group type;

[0139] A data splitting unit, used to split the spatiotemporal correlation data of the first user into a plurality of first spatiotemporal correlation data slices according to a preset time interval;

[0140] A slice clustering unit, used for performing density clustering on each first spatiotemporal correlation data slice in the same time interval to obtain a spatiotemporal clustering cluster in each time interval, wherein the time length of the time interval is equal to a preset time interval;

[0141] The cluster analysis unit is used to perform statistical analysis on the spatiotemporal clusters in each time interval to obtain a first group recognition result corresponding to the group type.

[0142] As an optional embodiment, before inputting the user portrait features of each user into a pre-trained user recognition model to obtain the first user corresponding to the group type, the user clustering module 220 may further include the following units:

[0143] A sample set acquisition unit is used to acquire a training sample set, the training sample set includes a plurality of training samples, each training sample includes a user portrait feature and a corresponding user type label, the user type label includes a user label belonging to a group type and a user label not belonging to a group type;

[0144] The model training unit is used to train the target model using the training samples until the training stop condition is met to obtain a user recognition model.

[0145] As an optional embodiment, the slice clustering unit specifically includes the following subunits:

[0146] The parameter setting subunit is used to set the model parameters of the initial clustering model to obtain the target clustering model;

[0147] A distance calculation subunit, used for calculating the interval distance between the position points corresponding to each first spatiotemporal correlation data slice according to each first spatiotemporal correlation data slice;

[0148] The slice clustering subunit is used to perform density clustering on each first spatiotemporal correlation data slice in the same time interval according to each interval distance through a target clustering model to obtain a spatiotemporal cluster in each time interval.

[0149] As an optional embodiment, the first spatiotemporal correlation data slice includes location point coordinates, and the location point coordinates include longitude coordinates, latitude coordinates, and altitude coordinates;

[0150] The distance calculation subunit is specifically used for:

[0151] Determine a horizontal distance between the first location point and the second location point according to the first longitude coordinate and the first latitude coordinate of the first location point, and the second longitude coordinate and the second latitude coordinate of the second location point, wherein the first location point and the second location point are location points corresponding to any two different first spatiotemporal correlation data slices;

[0152] The first altitude coordinate of the first location point is subtracted from the second altitude coordinate of the second location point and the absolute value is taken to obtain the altitude distance between the first location point and the second location point;

[0153] The interval distance between the first location point and the second location point is determined according to the horizontal distance and the altitude distance.

[0154] As an optional embodiment, the distance calculation subunit is further configured to:

[0155] The target cosine value, the cosine value of the first latitude coordinate, and the cosine value of the second latitude coordinate are multiplied to obtain a cosine calculation result, where the target cosine value is the cosine value of the difference between the first longitude coordinate and the second longitude coordinate;

[0156] Multiply the sine value of the first latitude coordinate and the sine value of the second latitude coordinate to obtain a sine calculation result;

[0157] The accumulated value of the cosine calculation result and the sine calculation result is arc-cosine-calculated and then multiplied by the target radius to obtain the horizontal distance between the first position point and the second position point.

[0158] As an optional embodiment, the slice clustering subunit is specifically used for:

[0159] Constructing a core point set according to the interval distances between the position points corresponding to the first spatiotemporal correlation data slices in the same time interval, wherein the core point set includes at least one core point among the position points corresponding to the first spatiotemporal correlation data slices;

[0160] The following steps are executed repeatedly until each core point in the core point set is traversed to obtain multiple spatiotemporal clusters in the time interval:

[0161] Select a target core point from the core point set, where the target core point is any core point that has not been selected;

[0162] The target core point is stored in the initial cluster, and the directly density reachable points corresponding to the target core point are stored in an ordered queue. The directly density reachable points in the ordered queue are sorted in ascending order according to the reachable distance between them and the target core point.

[0163] Obtain directly density-reachable points from the ordered queue in order, and determine whether the directly density-reachable points belong to core points;

[0164] When the directly density reachable point belongs to the core point, the target core point is updated to a directly density reachable point, and the loop execution is returned to store the target core point in the spatiotemporal cluster, and the directly density reachable points corresponding to the target core point are stored in the ordered queue until all the directly density reachable points in the ordered queue are traversed to obtain the spatiotemporal cluster in the time interval.

[0165] As an optional embodiment, the first group identification result includes first group user information, first group high frequency time period and first group high frequency area;

[0166] Cluster analysis unit, specifically used for:

[0167] Traversing the spatiotemporal clusters in each time interval, and determining the spatiotemporal clusters whose occurrence frequency is greater than or equal to a preset frequency threshold as frequent clusters;

[0168] Performing a union operation on each frequent cluster to obtain first group user information corresponding to the group type;

[0169] According to the time interval corresponding to each frequent cluster, the high-frequency time period of the first group corresponding to the group type is obtained by statistics;

[0170] Each frequent cluster is projected into the community after three-dimensional rasterization, and the high-frequency area of ​​the first group corresponding to the group type is obtained by statistics.

[0171] The user matching module 230 is specifically used for:

[0172] Splitting the spatiotemporal correlation data of each user in the community into a plurality of second spatiotemporal correlation data slices according to a preset time interval;

[0173] Matching the location points corresponding to each second spatiotemporal correlation data slice with the location of the target point of interest, and obtaining the number of occurrences and the frequency of occurrence of each user in the community at the target point of interest;

[0174] When the number of occurrences and the frequency of occurrences meet a preset statistical condition, determining that the user belongs to a second user corresponding to the group type;

[0175] Statistical analysis is performed on the spatiotemporal correlation data of each second user to obtain a second group recognition result corresponding to the group type.

[0176] Community-based identification method. Accordingly, the present application also provides a specific embodiment of a community-based identification device.

[0177] Figure 3 A schematic diagram of the hardware structure of a community group identification device provided in an embodiment of the present application is shown.

[0178] The community group identification device may include a processor 301 and a memory 302 storing computer program instructions.

[0179] Specifically, the processor 301 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0180] The memory 302 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 302 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 302 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 302 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 302 is a non-volatile solid-state memory.

[0181] The processor 301 reads and executes the computer program instructions stored in the memory 302 to implement any one of the community group identification methods in the above embodiments.

[0182] In one example, the community group identification device may further include a communication interface 303 and a bus 310. Figure 3 As shown, the processor 301, the memory 302, and the communication interface 303 are connected via a bus 310 and communicate with each other.

[0183] The communication interface 303 is mainly used to implement communication between each module, device, unit and / or equipment in the embodiment of the present application.

[0184] Bus 310 includes hardware, software or both, and the components of the identification device of the community group are coupled to each other. For example, but not limitation, the bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front-side bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 310 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the present application considers any suitable bus or interconnection.

[0185] In addition, in combination with the community group identification method in the above embodiment, the embodiment of the present application can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any community group identification method in the above embodiment is implemented.

[0186] In addition, in combination with the community group identification method in the above-mentioned embodiments, the embodiments of the present application may provide a computer program product for implementation. When the instructions in the computer program product are executed by the processor of an electronic device, the electronic device executes the community group identification method provided in any aspect of the above-mentioned embodiments of the present application.

[0187] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiment, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make each kind of change, modification and addition, or change the order between the steps after understanding the spirit of the present application.

[0188] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0189] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0190] Each aspect of the present disclosure is described above with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box 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, or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0191] The above is only a specific implementation of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of each equivalent modification or replacement within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application.

Claims

1. A method for identifying a community group, characterized in that: include: Obtaining user portrait features, spatiotemporal correlation data, and target points of interest in the community, where the target points of interest are used to characterize each point of interest corresponding to a group type of a target community group; Based on the user portrait features and the spatiotemporal correlation data, density clustering is performed on the users in the community to obtain a first group identification result corresponding to the group type; Matching the spatiotemporal association data of each user in the community with the location of the target point of interest to obtain a second group recognition result corresponding to the group type; The first group identification result and the second group identification result are integrated to obtain a target group identification result corresponding to the group type in the community.

2. The method according to claim 1, characterized in that The step of performing density clustering on the users in the community based on the user portrait features and the spatiotemporal association data to obtain a first group identification result corresponding to the group type includes: Inputting the user portrait features of each user into a pre-trained user recognition model to obtain a first user corresponding to the group type; Splitting the spatiotemporal correlation data of the first user into a plurality of first spatiotemporal correlation data slices according to preset time intervals; Density clustering is performed on each of the first spatiotemporal correlation data slices in the same time interval to obtain a spatiotemporal cluster in each of the time intervals, wherein the time length of the time interval is equal to the preset time interval; Statistical analysis is performed on the spatiotemporal clusters in each of the time intervals to obtain a first group recognition result corresponding to the group type.

3. The method according to claim 2, characterized in that Before inputting the user portrait features of each user into a pre-trained user recognition model to obtain a first user corresponding to the group type, the method further includes: Acquire a training sample set, the training sample set comprising a plurality of training samples, each of the training samples comprising the user portrait feature and its corresponding user type label, the user type label comprising a user label belonging to the group type and a user label not belonging to the group type; The target model is trained using the training samples until a training stop condition is met, thereby obtaining the user recognition model.

4. The method according to claim 2, characterized in that: The performing density clustering on each of the first spatiotemporal correlation data slices in the same time interval to obtain a spatiotemporal clustering cluster in each of the time intervals includes: The model parameters of the initial clustering model are set to obtain the target clustering model; Calculating the interval distances between the position points corresponding to the first spatiotemporal correlation data slices according to the first spatiotemporal correlation data slices; According to each of the interval distances, density clustering is performed on each of the first spatiotemporal correlation data slices in the same time interval using the target clustering model to obtain a spatiotemporal cluster in each of the time intervals.

5. The method according to claim 4, characterized in that The first spatiotemporal correlation data slice includes location point coordinates, and the location point coordinates include longitude coordinates, latitude coordinates and altitude coordinates; The calculating, according to each of the first spatiotemporal correlation data slices, the interval distance between the position points corresponding to each of the first spatiotemporal correlation data slices comprises: Determine, according to a first longitude coordinate and a first latitude coordinate of a first location point, and a second longitude coordinate and a second latitude coordinate of a second location point, a horizontal distance between the first location point and the second location point, wherein the first location point and the second location point are any two different location points corresponding to the first spatiotemporal association data slices; Subtract the first altitude coordinate of the first location point from the second altitude coordinate of the second location point and take the absolute value to obtain the altitude distance between the first location point and the second location point; The interval distance between the first location point and the second location point is determined according to the horizontal distance and the altitude distance.

6. The method according to claim 5, characterized in that The determining the horizontal distance between the first location point and the second location point according to the first longitude coordinate and the first latitude coordinate of the first location point, and the second longitude coordinate and the second latitude coordinate of the second location point, comprises: Multiplying a target cosine value, a cosine value of the first latitude coordinate, and a cosine value of the second latitude coordinate to obtain a cosine calculation result, wherein the target cosine value is a cosine value of a difference between the first longitude coordinate and the second longitude coordinate; Multiplying the sine value of the first latitude coordinate and the sine value of the second latitude coordinate to obtain a sine calculation result; The accumulated value of the cosine calculation result and the sine calculation result is arc-cosine-calculated and then multiplied by the target radius to obtain the horizontal distance between the first position point and the second position point.

7. The method according to claim 4, characterized in that The step of performing density clustering on each of the first spatiotemporal correlation data slices in the same time interval by using the target clustering model according to each of the interval distances to obtain a spatiotemporal cluster in each of the time intervals includes: Constructing a core point set according to the interval distances between the position points corresponding to the first spatiotemporal correlation data slices in the same time interval, wherein the core point set includes at least one core point among the position points corresponding to the first spatiotemporal correlation data slices; The following steps are executed in a loop until each of the core points in the core point set is traversed to obtain a plurality of spatiotemporal clusters in the time interval: Select a target core point from the core point set, wherein the target core point is any core point that has not been selected; The target core point is stored in an initial cluster, and the directly density reachable points corresponding to the target core point are stored in an ordered queue, and the directly density reachable points in the ordered queue are sorted in ascending order according to the reachable distance between them and the target core point; Obtaining the directly density-reachable points from the ordered queue in order, and determining whether the directly density-reachable points belong to the core points; In the case where the directly density reachable point belongs to the core point, the target core point is updated to the directly density reachable point, and the loop is returned to store the target core point in the spatiotemporal cluster, and the directly density reachable points corresponding to the target core point are stored in an ordered queue until all the directly density reachable points in the ordered queue are traversed to obtain the spatiotemporal cluster in the time interval.

8. The method according to claim 2, characterized in that: The first group identification result includes first group user information, first group high-frequency time period and first group high-frequency area; The performing statistical analysis on the spatiotemporal clusters in each of the time intervals to obtain a first group recognition result corresponding to the group type includes: Traversing the spatiotemporal clusters in each of the time intervals, and determining the spatiotemporal clusters whose occurrence frequency is greater than or equal to a preset frequency threshold as frequent clusters; Performing a union operation on each of the frequent clusters to obtain first group user information corresponding to the group type; According to the time interval corresponding to each of the frequent clusters, a first group high-frequency time period corresponding to the group type is obtained by statistics; Each of the frequent clusters is projected into the community after three-dimensional rasterization, and a first group high-frequency area corresponding to the group type is obtained by statistics.

9. The method according to any one of claims 1 to 8, characterized in that: The matching of the spatiotemporal association data of each user in the community with the location of the target point of interest to obtain a second group recognition result corresponding to the group type includes: Splitting the spatiotemporal correlation data of each user in the community into a plurality of second spatiotemporal correlation data slices according to preset time intervals; Matching the location points corresponding to each of the second spatiotemporal correlation data slices with the location of the target point of interest, and obtaining statistics on the number of times and frequency of occurrence of each user in the community at the target point of interest; In a case where the number of occurrences and the frequency of occurrence meet a preset statistical condition, determining that the user belongs to a second user corresponding to the group type; Statistical analysis is performed on the spatiotemporal correlation data of each of the second users to obtain a second group recognition result corresponding to the group type.

10. A device for identifying a community group, characterized in that: include: An information acquisition module, used to acquire user portrait features, spatiotemporal correlation data of users in the community, and target points of interest in the community, wherein the target points of interest are used to characterize each point of interest corresponding to a group type of a target community group; A user clustering module, configured to perform density clustering on users in the community based on the user portrait features and the spatiotemporal association data, to obtain a first group identification result corresponding to the group type; A user matching module, used for matching the spatiotemporal association data of each user in the community with the location of the target point of interest to obtain a second group recognition result corresponding to the group type; The result fusion module is used to fuse the first group identification result with the second group identification result to obtain a target group identification result corresponding to the group type in the community.

11. An electronic device, characterized in that: The device comprises: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the community group identification method as described in any one of claims 1-9.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the method for identifying a community group as described in any one of claims 1-9.

13. A computer program product, characterized in that When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the method for identifying a community group as described in any one of claims 1-9.

Images