User dwell point determination method and apparatus, electronic device, and computer storage medium

Abstract

The application relates to the technical field of mobile communication, and provides a user stay point determination method and device, electronic equipment and computer storage medium, the method comprises the following steps: determining travel base station data of a target user, and determining each target grid of the target user according to the travel base station data; determining the clustering center weight of each position point in each target grid according to the number of adjacent points and the moving speed of each position point in each target grid and the grid density of each target grid; determining each user stay grid of the target user according to the clustering center weight of each position point in each target grid, and determining the stay point of the target user according to each user stay grid. The user stay point determination method provided in the application embodiment takes the clustering center weight of each position point in each target grid as a determination standard, does not need to randomly select an initial clustering center in advance, reduces the randomness of the algorithm, and improves the accuracy of user stay point determination.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of mobile communication, in particular to a user stay point determination method and device, electronic equipment and computer storage medium. BACKGROUND

[0002] The existing user stay point determination mainly includes DBSCAN clustering method and K-Means clustering method. The K-Means clustering method is a typical distance-based non-hierarchical clustering algorithm, which divides the data into a predetermined number K on the basis of minimizing the error function, and uses distance as the evaluation index of similarity, that is, the closer the distance between two objects, the greater the similarity. The DBSCAN clustering method is a representative density-based clustering algorithm. Unlike partition and hierarchical clustering methods, it defines a cluster as the maximum set of points connected by density, and can divide areas with sufficiently high density into clusters and find clusters of arbitrary shape in noisy spatial databases.

[0003] The above-mentioned DBSCAN clustering method and K-Means clustering method cannot consider time series. Moreover, the DBSCAN clustering method and K-Means clustering method not only need to add distance threshold manually, but also need to set the number of clustering centers in advance, so the result may deviate from the global optimal classification. In actual operation, it is impossible to determine whether the clustering result is the optimal classification, so it is necessary to select different clustering centers and repeatedly run the algorithm, thereby making the time efficiency low, the randomness too large, and the accuracy of user stay point determination low. SUMMARY

[0004] The present application provides a user stay point determination method and device, electronic equipment and computer storage medium, aiming to improve the accuracy of user stay point determination.

[0005] In a first aspect, the present application provides a user stay point determination method, comprising:

[0006] determining the travel base station data of a target user, and determining each target grid of the target user according to the travel base station data;

[0007] determining the clustering center weight of each location point in each target grid according to the number of adjacent points and the moving speed of each location point in each target grid, and the grid density of each target grid;

[0008] determining each user stay grid of the target user according to the clustering center weight of each location point in each target grid, and determining the stay point of the target user according to each user stay grid.

[0009] In one embodiment, the cluster center weight of each location point in each of the target grids is determined according to the number of adjacent points and the moving speed of each location point in each of the target grids, and the grid density of each of the target grids, and the cluster center weight of each location point in each of the target grids is determined by:

[0010] The number of adjacent points and the moving speed of each location point in each of the target grids are normalized, and the grid density of each of the target grids is normalized to obtain the normalized value of the number of adjacent points and the moving speed of each location point in each of the target grids, and the normalized value of the grid density of each of the target grids;

[0011] The normalized value of the number of adjacent points and the moving speed of each location point in each of the target grids is multiplied by the normalized value of the grid density of each of the target grids to obtain the cluster center weight of each location point in each of the target grids.

[0012] The number of adjacent points and the moving speed of each location point in each of the target grids are determined, and the grid density of each of the target grids is determined by:

[0013] The length of stay of the target user in each of the target grids is determined according to the travel base station data, and the grid density of each of the target grids is determined according to the length of stay of each of the target grids;

[0014] Each location point in each of the target grids is taken as the center, and the number of adjacent points of each location point in each of the target grids is determined according to a preset distance threshold;

[0015] The moving speed of each location point in each of the target grids is determined according to the position distance and interval time of adjacent location points of each location point in each of the target grids.

[0016] The cluster center weight of each location point in each of the target grids is determined according to the cluster center weight of each location point in each of the target grids, and the cluster center weight of each location point in each of the target grids is determined by:

[0017] The cluster center weights of each location point in each of the target grids are summed to obtain the total cluster center weight of each of the target grids;

[0018] Based on the total cluster center weight of each of the target grids, the average cluster center weight of all target grids is calculated;

[0019] Based on the total cluster center weight of each of the target grids and the average cluster center weight, each of the user stay grids is determined.

[0020] The travel base station data of the target user is determined, and each of the target grids of the target user is determined according to the travel base station data, and the cluster center weight of each location point in each of the target grids is determined by:

[0021] determine the original base station data of a target user, and perform abnormal data screening on the original base station data to obtain travel base station data of the target user;

[0022] determine a target area according to longitude data and latitude data in the travel base station data, and rasterize the target area to obtain each grid of the target area;

[0023] map the travel base station data into each grid to determine each target grid of the target user.

[0024] The mapping of the travel base station data into each grid to determine each target grid of the target user comprises:

[0025] determine minimum longitude data and minimum latitude data in the travel base station data, and determine target longitude data and target latitude data of each location point in the travel base station data;

[0026] determine longitude span data and latitude span data of the grid of each location point;

[0027] determine first grid parameters of each location point according to the minimum longitude data, the target longitude data of each location point, and the longitude span data of the grid of each location point;

[0028] determine second grid parameters of each location point according to the minimum latitude data, the target latitude data of each location point, and the latitude span data of the grid of each location point;

[0029] map each location point into each grid according to the first grid parameters and the second grid parameters of each location point to determine each target grid.

[0030] The abnormal data screening on the original base station data to obtain the travel base station data of the target user comprises:

[0031] determine time stay intervals, location longitude data, and location latitude data of each adjacent location point in the original base station data;

[0032] calculate the moving distance of each adjacent location point according to the location longitude data and the location latitude data of each adjacent location point;

[0033] calculate the switching speed of each adjacent location point according to the moving distance and the time stay interval of each adjacent location point;

[0034] The original base station data is subjected to abnormal data screening according to the switching speed of each adjacent position point, and the original base station data after screening is subjected to position point filling to obtain the travel base station data.

[0035] The target area is determined according to the longitude data and the latitude data in the travel base station data, and the target area is rasterized to obtain each raster of the target area, including:

[0036] The maximum longitude data, the minimum longitude data, the maximum latitude data and the minimum latitude data in the travel base station data are determined.

[0037] The target area is determined according to the maximum longitude data, the minimum longitude data, the maximum latitude data and the minimum latitude data.

[0038] The area side length of the target area is determined, the longitude span of the target area is determined according to the area side length, the maximum longitude data and the minimum longitude data, and the latitude span of the target area is determined according to the area side length, the maximum latitude data and the minimum latitude data.

[0039] The target area is rasterized according to the area side length, the longitude span and the latitude span to obtain each raster of the target area.

[0040] In a second aspect, the present application provides a user stay point determination device, including:

[0041] A first determination module is configured to determine travel base station data of a target user, and determine each target raster of the target user according to the travel base station data.

[0042] A second determination module is configured to determine a clustering center weight of each position point in each target raster according to the number of adjacent points and the moving speed of each position point in each target raster, and the raster density of each target raster.

[0043] A determination and judgment module is configured to determine each user stay raster of the target user according to the clustering center weight of each position point in each target raster, and determine the stay point of the target user according to each user stay raster.

[0044] In a third aspect, the present application further provides an electronic device, including a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor executes the program to realize the user stay point determination method of the first aspect.

[0045] In a fourth aspect, the present application also provides a non-transitory computer readable storage medium comprising a computer program, which, when executed by the processor, implements the user stay point determination method of the first aspect.

[0046] In a fifth aspect, the present application also provides a computer program product comprising a computer program, which, when executed by the processor, implements the user stay point determination method of the first aspect.

[0047] The user stay point determination method, device, electronic device and computer storage medium provided by the present application use the clustering center weight of each position point in each target grid as a determination standard in the process of user stay point determination, do not need to randomly select an initial clustering center in advance, reduce the randomness of the algorithm, and improve the accuracy of user stay point determination. BRIEF DESCRIPTION OF DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the present application, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0049] Figure 1 is a flowchart of the user stay point determination method provided by the present application;

[0050] Figure 2 is a structural diagram of the user stay point determination device provided by the present application;

[0051] Figure 3 is a structural diagram of the electronic device provided by the present application. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the present application more clear, the technical solutions in the present application will be described clearly and completely in the following with reference to the drawings in the present application. Obviously, the described embodiments are some embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0053] In combination with Figures 1 to 3 The user stay point determination method, device, electronic device and computer storage medium provided by the present application are described. Figure 1 is a flowchart of the user stay point determination method provided by the present application; Figure 2 is a structural diagram of the user stay point determination device provided by the present application;Figure 3 is a structural schematic diagram of an electronic device provided by the present application.

[0054] Embodiments of the present application provide embodiments of a user stay point determination method. It should be noted that although a logical sequence is shown in the flowchart, under certain data, the steps shown or described can be completed in an order different from that shown.

[0055] Embodiments of the present application take an electronic device as an execution subject, and embodiments of the present application take a stay point determination system as one of the manifestations of the electronic device, and do not limit.

[0056] Referring to Figure 1 , Figure 1 is a flowchart of a user stay point determination method provided by the present application. The user stay point determination method provided by the embodiments of the present application comprises:

[0057] Step S10, determining the travel base station data of the target user, and determining each target grid of the target user according to the travel base station data.

[0058] It should be noted that most of the current research on stay point identification is based on user GPS trajectory data. Although the GPS trajectory data has high sampling density and accurate positioning accuracy, it depends on user active uploading and is not suitable for analyzing large-scale user travel flow analysis and trajectory identification. Therefore, the embodiments of the present application utilize base station data with passive user uploading, high credibility, easy acquisition, wide coverage and large data volume to identify the travel and stay points of users. It can be understood that the positions passed by the target user are covered by the base station, and the user terminal of the target user will automatically access the base station at the location.

[0059] Further, the stay point determination system needs to obtain the original base station data of the target user, and pre-process the original base station data to obtain the travel base station data of the target user. The travel base station data can be understood as the base station data of all position points passed by the target user. Further, the travel base station data includes but is not limited to user basic information, access base station date, user mobile phone number, access base station timestamp, access base station interval time, access base station longitude data and access base station latitude data.

[0060] Further, the stay point determination system determines each target grid passed by the target user according to the access base station longitude data and the access base station latitude data of each position point in the travel base station data, as described in steps S101 to S103.

[0061] In step S20, the clustering center weight of each position point in each target grid is determined according to the number of neighboring points and the moving speed of each position point in each target grid, and the grid density of each target grid.

[0062] It should be noted that the user stay point determination method of the embodiments of the present application is realized on the basis of the QMV model, and therefore, the embodiments of the present application can be further understood as a user stay point determination method based on the QMV model, wherein Q in the QMV model is the grid density of each target grid, M is the number of neighboring points of each position point in each target grid, and V is the moving speed of each position point in each target grid.

[0063] Further, the determination process of the grid density Q of each target grid, the number of neighboring points M of each position point in each target grid, and the moving speed V is as follows: for the grid density Q, the stay point determination system determines the stay duration of the target user in each target grid according to the travel base station data. The stay duration can be understood as the time difference between the start time of the start position point and the end time of the end position point in each target grid. The time unit can be minutes or hours, and the embodiments of the present application preferably use minutes as the time unit. That is, it can be understood as how many minutes the difference between the start time of the start position point and the end time of the end position point in each target grid is, and the grid density Q of each target grid is how many.

[0064] For the number of neighboring points M: first, a distance threshold D needs to be set, wherein the setting of the distance threshold D can be considered in combination with the following several cases. The first case is that the time interval is small and the distance is far apart, the user is in the middle of moving, and it is determined as a non-stay point. The second case is that the time interval is small and the distance is close, the user's mobile phone is in the situation of frequent dotting, and it needs to be determined according to the initial values of the time threshold and the distance threshold and the final result of clustering. The third case is that the time interval is large and the distance is far apart, the user can be in the middle of moving, and it needs to be determined according to the initial values of the time threshold and the distance threshold and the final result of clustering. The fourth case is that the time interval is large and the distance is close, the user is in non-moving, and it is determined as a stay point.

[0065] Further, the stay point determination system takes each position point of each target grid as the center and takes the distance threshold D as the preset distance threshold, and determines all position points within the distance threshold D as the neighboring points of each position point of each target grid. Further, the stay point determination system determines the number of neighboring points M of each position point in each target grid according to the number of neighboring points of each position point of each target grid. In an embodiment, the distance threshold D is 500 m (meters), a certain position point a in the target grid A, and the number of all position points within 500 m from the certain position point a is 10, that is, the number of neighboring points M of the certain position point a is 10.

[0066] For the moving speed V, the stay point discrimination system determines the first position distance and the second position distance of the adjacent position points of each position point in each target grid, and determines the first interval time and the second interval time of the adjacent position points of each position point in each target grid. Then, the stay point discrimination system sums the first position distance and the second position distance of the adjacent position points of each position point in each target grid to obtain the position distance sum value of each position point in each target grid. Next, the stay point discrimination system sums the first interval time and the second interval time of the adjacent position points of each position point in each target grid to obtain the interval time sum value of each position point in each target grid. Finally, the stay point discrimination system divides the position distance sum value of each position point in each target grid by the interval time sum value to obtain the moving speed of each position point in each target grid.

[0067] It is further needed to be explained that, since the start position point and the end position point in each target grid only have one adjacent position point, the moving speed of the start position point and the end position point in each target grid is actually the quotient value of the position distance and the interval time of the start position point and the end position point in each target grid and the one adjacent position point.

[0068] Further, the stay point discrimination system normalizes the number of adjacent points of each position point in each target grid to obtain a first normalization value, normalizes the moving speed of each position point in each target grid to obtain a second normalization value, and normalizes the grid density of each target grid to obtain a third normalization value. Finally, the stay point discrimination system calculates the cluster center weight value of each position point in each target grid according to the first normalization value, the second normalization value and the third normalization value, which is specifically described in steps S201 to S202.

[0069] Step S30, according to the cluster center weight value of each position point in each target grid, determines each user stay grid of the target user, and discriminates the stay point of the target user according to each user stay grid.

[0070] Further, the stay point determination system sums the cluster center weights of each position point in each target grid to obtain the total cluster center weight of each target grid. Then, the stay point determination system calculates the average cluster center weight of all target grids according to the total cluster center weight of each target grid and the number of target grids. Next, the stay point determination system compares the total cluster center weight of each target grid with the average cluster center weight to obtain a comparison result, where the comparison result can be that the total cluster center weight of a target grid is greater than or equal to the average cluster center weight, or the comparison result can be that the total cluster center weight of a target grid is less than the average cluster center weight. Finally, the stay point determination system determines the user stay grids according to the comparison result, as described in steps S301 to S303.

[0071] Further, the stay point determination system determines the specific location information of each user stay grid and determines the stay point of the target user according to the specific location information. In an embodiment, the stay point can depict the travel habits of the target user, determine the residence and workplace of the target user, and further determine the occupation type of the target user. At the same time, the stay point of the target user on holidays can be projected to the actual geographical location, such as a shopping mall, a restaurant, or the like, and the consumption level and consumption preferences of the target user can be depicted through the actual location where the target user often stays. In terms of epidemic prevention, the stay points of a large number of target users can be tracked to determine the flow of people in a certain area, and the crowd can be guided and the epidemic prevention work can be performed again.

[0072] The embodiment provides a user stay point determination method, which takes the cluster center weight of each position point in each target grid as a determination standard in the process of determining the user stay point, does not need to randomly select an initial cluster center in advance, reduces the randomness of the algorithm, and improves the accuracy of determining the user stay point.

[0073] Further, steps S101 to S103 are described as follows:

[0074] Step S101, determining the original base station data of a target user, and removing abnormal data from the original base station data to obtain travel base station data of the target user;

[0075] Step S102, determining a target area according to the longitude data and the latitude data in the travel base station data, and rasterizing the target area to obtain each grid of the target area;

[0076] Step S103, mapping the travel base station data to each grid to determine each target grid of the target user.

[0077] Specifically, the stay point determination system determines the original base station data of the target user. Considering that the accuracy of the base station data is not as good as that of the GPS data, and there is a certain drift, oscillation, loss and deviation, the stay point determination system needs to filter and delete the abnormal data of the original base station data. Due to the overlapping coverage of the base station, the position switching and the base station load, the dotting data of the target user on the base station may be repeated, and the distance difference is too large in a short time, which leads to abnormality. The characteristics of abnormal data are often rapid repeated movement of the trajectory in a short time, large change distance of the trajectory in a short time, and fast switching speed. Therefore, the switching speed of each position point of the target user is needed to filter and delete the abnormal data of the original base station data, so as to obtain the travel base station data of the target user, which is specifically described in steps S1011 to S1014.

[0078] Further, the stay point determination system determines the longitude data and latitude data in the travel base station data, wherein the longitude data and latitude data here are the access base station longitude data and access base station latitude data described in step S10. The longitude data includes but is not limited to the maximum longitude data and the minimum longitude data, and the latitude data includes but is not limited to the maximum latitude data and the minimum latitude data. The stay point determination system determines the target area where the target user is located in the whole process according to the maximum longitude data and the minimum longitude data, and further rasterizes the target area to obtain each grid of the target area, which is specifically described in steps S1021 to S1024.

[0079] Further, the stay point determination system determines each grid where each position point is located according to the longitude data and latitude data of each position point in the travel base station data, and the maximum longitude data, the minimum longitude data, the maximum latitude data and the minimum latitude data. It can be understood that each position point is mapped into each grid to determine each target grid of the target user, which is specifically described in steps S1031 to S1035.

[0080] The embodiment of the present application maps each position point passed by the target user into each grid to more accurately determine the area passed by the target user.

[0081] Further, the steps S1011 to S1014 are described as follows:

[0082] Step S1011, determining the time stay interval, position longitude data and position latitude data of each adjacent position point in the original base station data;

[0083] Step S1012, calculating the moving distance of each adjacent position point according to the position longitude data and position latitude data of each adjacent position point;

[0084] Step S1013, according to the moving distance and the time interval of each adjacent position point, the switching speed of each adjacent position point is calculated;

[0085] Step S1014, according to the switching speed of each adjacent position point, the abnormal data of the original base station data is screened out, and the position point filling is performed on the screened original base station data, so as to obtain the travel base station data.

[0086] Specifically, the stay point discrimination system determines the time interval, the position longitude data and the position latitude data of each adjacent position point in the original base station data, wherein the position longitude data and the position latitude data are also the access base station longitude data and the access base station latitude data described in step S10. Then, the stay point discrimination system calculates the moving distance of each adjacent position point according to the position longitude data and the position latitude data of each adjacent position point. Then, the stay point discrimination system obtains the switching speed of each adjacent position point according to the moving distance of each adjacent position point divided by the time interval of each adjacent position point.

[0087] Further, the stay point discrimination system determines the preset speed threshold, and compares the switching speed of each adjacent position point with the speed threshold to obtain the comparison result, wherein the comparison result can be that the switching speed of a certain adjacent position point is greater than or equal to the speed threshold, and the comparison result can also be that the switching speed of a certain adjacent position point is less than the speed threshold.

[0088] Further, if it is determined that the comparison result is that the switching speed of a certain adjacent position point is greater than or equal to the speed threshold, the stay point discrimination system determines that the switching speed is too large, determines that the switching position point has a shock or abnormal drift of position record, that is, the switching position point is an abnormal position point, and all switching position points whose switching speed is greater than or equal to the speed threshold are removed. In an embodiment, the speed threshold is 80 m / s (meter / second), and the switching speed of a certain adjacent position point a and b is 500 m / s, wherein b is a switching position point, since the switching speed 500 m / s is greater than the speed threshold 80 m / s, it is determined that the switching position point b is an abnormal position point, and the switching position point b is removed. Further, the base station data after removing the abnormal position point obviously identifies two continuous trajectories of the target user.

[0089] Further, the screened original base station data has the problems of missing base station position and discontinuous position points of the target user in time. In order to obtain more accurate travel base station data and reduce the influence of missing values, the screened original base station data also needs to be filled with position points. The embodiments of the present application preferably fill the position points by using a data interpolation method based on time and space thresholds, and the specific process is as follows: when the time interval is not large and the target user moves a certain distance according to the base station data of adjacent position points in such a missing time period, it is determined that there is a missing position point between the adjacent position points. Finally, the missing time of the missing time point is divided by the pre-defined missing time interval n to obtain the number of interpolations, and the adjacent position points are filled equidistantly to obtain the travel base station data. The specific process of filling the adjacent position points equidistantly is as follows:

[0090] Further, the stop point discrimination system sorts the obtained target user position sequence according to time, and calculates the base station distance and time interval of adjacent position points of the target user. Then, the stop point discrimination system judges according to the distance threshold and the time threshold. In an embodiment, the missing position points with a moving distance less than 500 m and a missing time less than 30 min are selected for equidistant interpolation. It can be further understood that if the base station distance of adjacent position points is less than 500 m and the missing time is less than 30 min, the stop point discrimination system inserts a new position point between the adjacent position points. The longitude and latitude data of the new position point is the average of the longitude and latitude data of the adjacent position points. It should be noted that the adjacent position points in the embodiments of the present application can be described as the previous position point and the next position point, the start time is recorded as St, and the end time is recorded as Et. The original base station data has been processed, and the start time Et1 of the next position point is equal to the end time St2 of the previous position point. Referring to Table 1 and Table 2, Table 1 is the original base station data, and Table 2 is the interpolated original base station data.

[0091] Table 1 original base station data

[0092] Location point Longitude data Latitude data Start time End time Previous location point Lon1 Lat1 St1 Et1 Next location point Lon2 Lat2 St2 Et2

[0093] Table 2 interpolated original base station data

[0094] Location point Longitude data Latitude data Start time End time Previous location point Lon1 Lat1 St1 St1 + (St2 - St1) / 2 New location point (Lon1 + Lon2) / 2 (Lat1 + Lat2) / 2 St1 + (St2 - St1) / 2 St2 Next location point Lon2 Lat2 St2 Et2

[0095] Further, if the time interval exceeds 30 min, the stop point discrimination system fills the adjacent position points equidistantly according to the number of interpolations of the missing time divided by the pre-defined missing time interval n, wherein lon1 is less than lon2 by default.

[0096] In an embodiment, if the time interval is less than 30 minutes, the distance between the front and rear connected base stations is less than 500 m, the end time of the previous position point in the adjacent position point is 10 o'clock, the start time of the next position point in the adjacent position point is 12 o'clock, and the interval time between the previous position point and the next position point in the adjacent position point is 2 hours, that is, 120 minutes, the number of records required for interpolation n = (120 / 30)-1 = 3, that is, 3 new position points are required to be filled at equal intervals, and the 3 new position points are new position point 1, new position point 2 and new position point 3. Referring to Table 3 and Table 4, Table 3 is the original base station data, and Table 4 is the original base station data after interpolation.

[0097] Table 3 Original base station data

[0098] Location point Longitude data Latitude data Start time End time Previous location point Lon1 Lat1 St1 Et1 Next location point Lon2 Lat2 St2 Et2

[0099] Table 4 Original base station data after interpolation

[0100]

[0101] The embodiments of the present application can obtain the continuous travel trajectory of the target user by screening out abnormal data and filling position points from the original base station data, and make the travel trajectory more coherent, and can more truly fit the actual motion trajectory of the target user.

[0102] Further, steps S1021 to S1024 are described as follows:

[0103] Step S1021, determining the maximum longitude data, the minimum longitude data, the maximum latitude data and the minimum latitude data in the travel base station data;

[0104] Step S1022, determining the target area according to the maximum longitude data, the minimum longitude data, the maximum latitude data and the minimum latitude data;

[0105] Step S1023, determining the area side length of the target area, determining the longitude span of the target area according to the area side length, the maximum longitude data and the minimum longitude data, and determining the latitude span of the target area according to the area side length, the maximum latitude data and the minimum latitude data;

[0106] Step S1024, rasterizing the target area according to the area side length, the longitude span and the latitude span, to obtain each grid of the target area.

[0107] Specifically, the stay point determination system determines the maximum longitude data, the minimum longitude data, the maximum latitude data and the minimum latitude data in the travel base station data according to the longitude data and the latitude data of each position point in the travel base station data. Then, the stay point determination system determines the maximum longitude extreme point, the minimum longitude extreme point, the maximum latitude extreme point and the minimum latitude extreme point according to the maximum longitude data, the minimum longitude data, the maximum latitude data and the minimum latitude data. Then, the stay point determination system determines the original target area passed by the target user according to the maximum longitude extreme point, the minimum longitude extreme point, the maximum latitude extreme point and the minimum latitude extreme point in the travel base station data.

[0108] In an embodiment, the maximum longitude data is 119.22245, the minimum longitude data is 118.360001, the maximum latitude data is 32.60741 and the minimum latitude data is 31.23242, thus, the maximum longitude extreme point is (119.22245, 32.60741), the minimum longitude extreme point is (118.360001, 31.23242), the maximum latitude extreme point is (118.360001, 32.60741) and the minimum latitude extreme point is (119.22245, 31.23242), further, the maximum longitude extreme point is the upper right corner position point, the minimum longitude extreme point is the lower left corner position point, the maximum latitude extreme point is the upper left corner position point and the minimum latitude extreme point is the lower right corner position point.

[0109] Further, the stay point determination system determines the area width and the area length of the original target area, if the area width is greater than the area length, the area width is determined as the area side length of the original target area, if the area width is less than the area length, the area length is determined as the area side length of the original target area. Then, the stay point determination system determines the side length of the minimum circumscribed square by taking the area side length of the original target area by a preset length upwards, wherein the preset length is set according to the actual situation, which is not limited in the embodiment. Then, the stay point determination system determines the lower left corner position point as the origin position point, the upper right corner position point, the upper left corner position point and the lower right corner position point extend outward by the side length of the minimum circumscribed square, to construct the minimum circumscribed square, and determines the minimum circumscribed square as the final target area.

[0110] Further, the stay point discrimination system calculates the difference between the maximum longitude data and the minimum longitude data to obtain the original longitude span, and then determines the extended longitude data according to the region side length of the final target region and the region length of the original target region, and calculates the sum of the extended longitude data and the original longitude span to obtain the longitude span of the final target region. Next, the stay point discrimination system calculates the difference between the maximum latitude data and the minimum latitude data to obtain the original latitude span, and then determines the extended latitude data according to the region side length of the final target region and the region width of the original target region, and calculates the sum of the extended latitude data and the original latitude span to obtain the latitude span of the final target region.

[0111] Further, the stay point discrimination system rasterizes the final target region according to the region side length, the longitude span and the latitude span of the final target region to obtain each grid of the final target region. In an embodiment, the maximum longitude data of the original target region is 119.22245, the minimum longitude data is 118.360001, the maximum latitude data is 32.60741, and the minimum latitude data is 31.23242, the region width of the original target region is 152899.071 m, the region length is 81241.876 m, and the preset length is 100 m. Therefore, the longitude span of the original target region is calculated as 119.22245-118.360001=0.86244, the latitude span of the original target region is calculated as 32.60741-31.23242=1.37499, the region side length of the original target region is 152899.071 m, and after taking 100 m upward, it is 153000 m as the region side length of the final target region. Therefore, according to the region side length 153000 m and the region width 152899.071 m, the extended latitude data is determined as 0.00092, the latitude span of the final target region is 1.37591, according to the region side length 153000 m and the region length 81241.876 m, the extended longitude data is determined as 0.76178, and the longitude span of the final target region is 1.62422. Further, the final target region is rasterized according to the region side length 153000 m, the longitude span 1.62422 and the latitude span 1.37591 of the final target region, to obtain each grid of the final target region. It can be understood that 1 longitude=153000 m / 1.62422=94199.05 m, and 1 latitude=153000 m / 1.37591=111198.92 m. As shown in Table 5, Table 5 is the grid of the final target region.

[0112] Table 5 Grid of the final target region

[0113]

[0114] Further, the stay point determination system determines the base station area type of each grid in the final target area, and according to the number of base station area types of each grid in the final target area, the stay point determination system further subdivides each grid in the final target area, wherein the base station area type includes but is not limited to a suburb type, an urban type, and a main urban type.

[0115] Further, if the most number of base station area types of each grid in the final target area is determined to be the suburb type, the stay point determination system further divides each grid in the final target area into a first preset size, wherein the first preset size is the original grid size, and the original grid size can be set according to the actual situation of the specific location of the final target area, for example, the original grid size is 1000m*1000m.

[0116] Further, if the most number of base station area types of each grid in the final target area is determined to be the urban type, the stay point determination system further divides each grid in the final target area into a second preset size, wherein the second preset size is half of the first preset size, that is, the second preset size is half of the original grid size, for example, the original grid size is 1000m*1000m, and the second preset size is 500m*500m.

[0117] Further, if the most number of base station area types of each grid in the final target area is determined to be the main urban type, the stay point determination system further divides each grid in the final target area into a third preset size, wherein the third preset size is 1 / 10 of the second preset size, that is, the third preset size is 1 / 20 of the original grid size, for example, the original grid size is 1000m*1000m, and the third preset size is 50m*50m.

[0118] The embodiments of the present application accurately grid the target area by the area side length, the longitude span, and the latitude span, and further refine the size of each grid according to the base station area type of each grid, so as to obtain more accurate grids of the target area.

[0119] Further, the steps S1031 to S1035 are described as follows:

[0120] In step S1031, the minimum longitude data and the minimum latitude data in the travel base station data are determined, and the target longitude data and the target latitude data of each location point in the travel base station data are determined.

[0121] In step S1032, the longitude span data and the latitude span data of the grid of each location point are determined.

[0122] Step S1033: Determine the first grid parameters of each location point based on the minimum longitude data, the target longitude data of each location point, and the longitude span data of the grid at each location point.

[0123] Step S1034: Determine the second grid parameters for each location point based on the minimum latitude data, the target latitude data for each location point, and the latitude span data of the grid for each location point.

[0124] Step S1035: Based on the first grid parameter and the second grid parameter of each location point, map each location point to a grid to determine each target grid.

[0125] Specifically, the stop point determination system determines the minimum longitude data (lon) in the travel base station data. min and minimum latitude data lat min It also determines the target longitude data (lon) and target latitude data (lat) of each location point in the travel base station data, and simultaneously determines the longitude span data (lon) of the grid at each location point. Eps Latitude span data lat Eps Next, the stop point determination system uses the roundup() method to round up the data, combined with the minimum longitude data (lon). min The target longitude data (lon) for each location point and the longitude span data (lon) of the raster at each location point. Eps Determining the first grid parameter x for each location point can be expressed as: Furthermore, the stop point identification system uses the roundup() rounding up method, combined with the minimum latitude data lat min The target latitude data (lat) for each location point and the latitude span data (lat) of the raster at each location point. Eps Determine the second grid parameter y for each location point. The second grid parameter y can be expressed as:

[0126] Furthermore, the dwell point discrimination system maps each location point to a grid based on each first grid parameter x and each second grid parameter y, thereby determining each target grid. Specifically, the dwell point discrimination system represents the first grid parameter x and each second grid parameter y as (x, y), where x represents the x-th grid in the horizontal coordinate direction, y represents the y-th grid in the vertical coordinate direction, and (x, y) represents the grid where the horizontal and vertical coordinates intersect, thus mapping each location point to a grid and determining the target grid for each location point.

[0127] In this embodiment, each location point is mapped to a grid, and each target grid is determined, thereby accurately determining the grid where each location point is located.

[0128] Further, the steps S201 to S202 are described as follows:

[0129] In step S201, the number of neighboring points and the moving speed of each position point in each target grid are normalized, and the grid density of each target grid is normalized, to obtain the normalized value of the number of neighboring points and the moving speed of each position point in each target grid, and the normalized value of the grid density of each target grid.

[0130] In step S202, the normalized value of the number of neighboring points and the moving speed of each position point in each target grid is multiplied by the normalized value of the grid density of each target grid, to obtain the cluster center weight of each position point in each target grid.

[0131] Specifically, for the normalization of the number of neighboring points, the stay point determination system determines the maximum number of neighboring points and the minimum number of neighboring points in each target grid, and records the maximum number of neighboring points as Max M , and records the minimum number of neighboring points as Min M . Then, the stay point determination system normalizes the number of neighboring points M M of each position point in each target grid according to the maximum number of neighboring points Max M , the minimum number of neighboring points Min i , and the number of neighboring points M i of each position point in each target grid, to obtain the first normalized value of each position point in each target grid, which can be expressed as

[0132] Further, for the normalization of the moving speed, the stay point determination system determines the maximum speed and the minimum speed of each target grid, and records the maximum speed as Max V , and records the minimum speed as Min V . Then, the stay point determination system normalizes the moving speed of each position point in each target grid according to the maximum speed Max V , the minimum speed Min V , and the moving speed V i of each position point in each target grid, to obtain the second normalized value of each position point in each target grid, which can be expressed as

[0133] Further, for the normalization of the target grids: the stay point determination system determines the target grid with the maximum grid density and the target grid with the minimum grid density in each target grid, and records the target grid with the maximum grid density as Max Q , and records the target grid with the minimum grid density as Min Q . Then, the stay point determination system normalizes the grid density of each target grid according to the target grid with the maximum grid density Max Q , the target grid with the minimum grid density Min Q , and the grid density Q i of each target grid, to obtain a third normalized value of each target grid, which can be expressed as

[0134] Further, the stay point determination system multiplies the first normalized value of each position point in each target grid, the second normalized value of each position point in each target grid, and the third normalized value of each target grid, to obtain the cluster center weight of each position point in each target grid.

[0135] In the process of user stay point determination, the embodiments of the present application take the cluster center weight of each position point in each target grid as the determination standard, without the need to randomly select an initial cluster center in advance, thereby reducing the randomness of the algorithm and improving the accuracy of user stay point determination.

[0136] Further, the steps S301 to S303 are described as follows:

[0137] Step S301: summing the cluster center weights of each position point in each target grid to obtain the cluster center total weight of each target grid;

[0138] Step S302: calculating the cluster center average weight of all target grids based on the cluster center total weight of each target grid;

[0139] Step S303: determining each user stay grid based on the cluster center total weight of each target grid and the cluster center average weight.

[0140] Specifically, the dwell point identification system sums the cluster center weights of each location point in each target grid to obtain the total cluster center weight of each target grid. Next, based on the total cluster center weights of each target grid and the number of target grids, the system calculates the average cluster center weight of all target grids. Then, the system compares the total cluster center weights of each target grid with the average cluster center weight, obtaining a comparison result. This result can be that the total cluster center weight of a target grid is greater than or equal to the average cluster center weight, or it can be that the total cluster center weight of a target grid is less than the average cluster center weight. If it is determined that the total cluster center weight of a target grid is greater than or equal to the average cluster center weight, the system identifies that target grid as the user's dwell grid for the target user; that is, all target grids with a total cluster center weight greater than or equal to the average cluster center weight are identified as the user's dwell grid for the target user.

[0141] In the process of user dwell point determination, the cluster center weights of each location point in each target grid are used as the determination criteria. This eliminates the need to randomly select initial cluster centers in advance, reducing the randomness of the algorithm and improving the accuracy of user dwell point determination.

[0142] Furthermore, the user dwell point determination device provided in this application will be described below. The user dwell point determination device and the user dwell point determination method can be referred to each other accordingly.

[0143] like Figure 2 As shown, Figure 2 This is a schematic diagram of the user dwell point determination device provided in this application. The user dwell point determination device includes:

[0144] The first determining module 201 is used to determine the travel base station data of the target user and determine each target grid of the target user based on the travel base station data;

[0145] The second determining module 202 is used to determine the cluster center weight of each location point in each target grid based on the number of neighboring points and the moving speed of each location point in each target grid, as well as the grid density of each target grid.

[0146] The determination and discrimination module 203 is used to determine each user stay grid of the target user based on the cluster center weight of each location point in each target grid, and to determine the stay point of the target user based on each user stay grid.

[0147] Furthermore, the first determining module 201 is also used for:

[0148] determine original base station data of a target user, and perform abnormal data screening on the original base station data to obtain travel base station data of the target user;

[0149] determine a target area according to longitude data and latitude data in the travel base station data, and rasterize the target area to obtain each grid of the target area;

[0150] map the travel base station data to each of the grids to determine each target grid of the target user.

[0151] Further, the first determining module 201 comprises a mapping unit, configured to:

[0152] determine minimum longitude data and minimum latitude data in the travel base station data, and determine target longitude data and target latitude data of each location point in the travel base station data;

[0153] determine longitude span data and latitude span data of a grid of each of the location points;

[0154] determine first grid parameters of each of the location points according to the minimum longitude data, the target longitude data of each of the location points, and the longitude span data of the grid of each of the location points;

[0155] determine second grid parameters of each of the location points according to the minimum latitude data, the target latitude data of each of the location points, and the latitude span data of the grid of each of the location points;

[0156] map each of the location points to each grid according to the first grid parameters and the second grid parameters of each of the location points to determine each of the target grids.

[0157] Further, the first determining module 201 comprises a screening and filling unit, configured to:

[0158] determine time stay intervals, location longitude data and location latitude data of each adjacent location point in the original base station data;

[0159] calculate moving distances of each of the adjacent location points according to the location longitude data and the location latitude data of each of the adjacent location points;

[0160] calculate switching speeds of each of the adjacent location points according to the moving distances and the time stay intervals of each of the adjacent location points;

[0161] perform abnormal data screening on the original base station data according to the switching speeds of each of the adjacent location points, and perform location point filling on the screened original base station data to obtain the travel base station data.

[0162] Further, the first determining module 201 comprises a gridding unit, configured to:

[0163] determine maximum longitude data, minimum longitude data, maximum latitude data and minimum latitude data in the travel base station data;

[0164] determine the target area according to the maximum longitude data, the minimum longitude data, the maximum latitude data and the minimum latitude data;

[0165] determine the area side length of the target area, determine the longitude span of the target area according to the area side length, the maximum longitude data and the minimum longitude data, and determine the latitude span of the target area according to the area side length, the maximum latitude data and the minimum latitude data;

[0166] gridding the target area according to the area side length, the longitude span and the latitude span, to obtain each grid of the target area.

[0167] Further, the second determining module 202 is further configured to:

[0168] normalize the number of adjacent points and the moving speed of each position point in each target grid, and the grid density of each target grid, to obtain the normalized value of the number of adjacent points and the moving speed of each position point in each target grid, and the normalized value of the grid density of each target grid;

[0169] multiply the normalized value of the number of adjacent points and the moving speed of each position point in each target grid, and the normalized value of the grid density of each target grid, to obtain the clustering center weight of each position point in each target grid.

[0170] Further, the second determining module 202 is further configured to:

[0171] determine the stay duration of the target user in each target grid according to the travel base station data, and determine the grid density of each target grid according to the stay duration of each target grid;

[0172] take each position point in each target grid as the center, and determine the number of adjacent points of each position point in each target grid according to a preset distance threshold;

[0173] determine the moving speed of each position point in each target grid according to the position distance and interval time of adjacent position points of each position point in each target grid.

[0174] Further, the determination and discrimination module 203 is further configured to:

[0175] The cluster center weights of each location point in each target grid are summed to obtain the total cluster center weight of each target grid.

[0176] Based on the total weight of the cluster centers of each target grid, the average weight of the cluster centers of all target grids is calculated.

[0177] Each user stay grid is determined based on the total weight of the cluster centers and the average weight of the cluster centers of each target grid.

[0178] The specific embodiments of the user dwell point determination device provided in this application are basically the same as the embodiments of the user dwell point determination method described above, and will not be repeated here.

[0179] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3 As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330, and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute a user dwell point determination method, which includes:

[0180] Determine the travel base station data of the target user, and determine each target grid of the target user based on the travel base station data;

[0181] Based on the number of neighboring points and the moving speed of each location point in each target grid, as well as the grid density of each target grid, the cluster center weight of each location point in each target grid is determined.

[0182] Based on the cluster center weights of each location point in each target grid, the user stay grids of the target user are determined, and the stay points of the target user are identified based on each user stay grid.

[0183] Further, the logic instructions in the memory 330 described above can be implemented in the form of software functional units and sold or used as independent products, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or parts of the present application that essentially contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0184] In another aspect, the present application also provides a computer program product, which comprises a computer program stored on a non-transitory computer readable storage medium, and the computer program comprises program instructions, and when the program instructions are executed by a computer, the computer can execute the user stay point determination method provided by the above-mentioned methods, and the method comprises:

[0185] determining travel base station data of a target user, and determining each target grid of the target user according to the travel base station data;

[0186] determining a cluster center weight of each location point in each target grid according to the number of adjacent points and the moving speed of each location point in each target grid, and the grid density of each target grid;

[0187] determining each user stay grid of the target user according to the cluster center weight of each location point in each target grid, and determining the stay point of the target user according to each user stay grid.

[0188] In still another aspect, the present application also provides a non-transitory computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the user stay point determination method provided by the above-mentioned methods, and the method comprises:

[0189] determining travel base station data of a target user, and determining each target grid of the target user according to the travel base station data;

[0190] determining a cluster center weight of each location point in each target grid according to the number of adjacent points and the moving speed of each location point in each target grid, and the grid density of each target grid;

[0191] According to the clustering center weight of each position point in each target grid, each user stay grid of the target user is determined, and a stay point of the target user is determined according to each user stay grid.

[0192] The device embodiments described above are only illustrative, wherein the units described as separate components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0193] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0194] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A user stay point determination method characterized by comprising: The method comprises the following steps: determining travel base station data of a target user, and determining target grids of the target user according to the travel base station data; determining a clustering center weight of each location point in each target grid according to the number of adjacent points and the moving speed of each location point in each target grid, and the grid density of each target grid, comprising: normalizing the number of adjacent points and the moving speed of each location point in each target grid, and the grid density of each target grid, to obtain the normalized values of the number of adjacent points and the moving speed of each location point in each target grid, and the normalized values of the grid density of each target grid; multiplying the normalized values of the number of adjacent points and the moving speed of each location point in each target grid, and the normalized values of the grid density of each target grid, to obtain the clustering center weight of each location point in each target grid; determining user stay grids of the target user according to the clustering center weight of each location point in each target grid, and determining stay points of the target user according to specific location information of each user stay grid; the step of determining the user stay grids of the target user according to the clustering center weight of each location point in each target grid comprises: summing the clustering center weight of each location point in each target grid to obtain a total clustering center weight of each target grid; calculating an average clustering center weight of all target grids based on the total clustering center weight of each target grid; and determining each user stay grid based on the total clustering center weight of each target grid and the average clustering center weight.

2. The user dwell point discrimination method of claim 1, wherein determining the number of adjacent points and the moving speed of each location point in each target grid, and the grid density of each target grid, comprises: determining the stay duration of the target user in each target grid according to the travel base station data, and determining the grid density of each target grid according to the stay duration of each target grid; determining the number of adjacent points of each location point in each target grid based on the location points of each target grid and a preset distance threshold; determining the moving speed of each location point in each target grid based on the location distance and interval time of adjacent location points of each location point in each target grid.

3. The user dwell point discrimination method of any of claims 1-2, wherein, the step of determining the travel base station data of a target user, and determining target grids of the target user according to the travel base station data, comprises: determining original base station data of a target user, and screening out abnormal data from the original base station data to obtain travel base station data of the target user; determining a target area according to the longitude data and the latitude data in the travel base station data, and rasterizing the target area to obtain each grid of the target area; mapping the travel base station data to each grid to determine target grids of the target user.

4. The user dwell point discrimination method of claim 3, wherein the step of mapping the travel base station data to each grid to determine target grids of the target user, comprises: determine minimum longitude data and minimum latitude data in the travel base station data, and determine target longitude data and target latitude data of each location point in the travel base station data; determine longitude span data and latitude span data of a grid of each of the location points; determine first grid parameters of each of the location points according to the minimum longitude data, the target longitude data of each of the location points, and the longitude span data of the grid of each of the location points; determine second grid parameters of each of the location points according to the minimum latitude data, the target latitude data of each of the location points, and the latitude span data of the grid of each of the location points; map each of the location points into each grid according to the first grid parameters and the second grid parameters of each of the location points, and determine each of the target grids.

5. The user dwell point discrimination method of claim 3, wherein, The abnormal data screening on the original base station data to obtain the travel base station data of the target user comprises: determine time interval of stay, location longitude data and location latitude data of each adjacent location point in the original base station data; calculate moving distance of each of the adjacent location points according to the location longitude data and the location latitude data of each of the adjacent location points; calculate switching speed of each of the adjacent location points according to the moving distance and the time interval of stay of each of the adjacent location points; perform abnormal data screening on the original base station data according to the switching speed of each of the adjacent location points, and perform location point filling on the screened original base station data to obtain the travel base station data.

6. The user dwell point discrimination method of claim 3, wherein, The determination of the target region according to the longitude data and the latitude data in the travel base station data, and the gridding of the target region to obtain each grid of the target region, comprises: determine maximum longitude data, minimum longitude data, maximum latitude data and minimum latitude data in the travel base station data; determine the target region according to the maximum longitude data, the minimum longitude data, the maximum latitude data and the minimum latitude data; determine region side length of the target region, determine longitude span of the target region according to the region side length, the maximum longitude data and the minimum longitude data, and determine latitude span of the target region according to the region side length, the maximum latitude data and the minimum latitude data; perform gridding on the target region according to the region side length, the longitude span and the latitude span to obtain each grid of the target region.

7. A user stay point determination device characterized by comprising: comprise: a first determination module configured to determine travel base station data of a target user, and determine each target grid of the target user according to the travel base station data; The second determining module is configured to determine the clustering center weight of each position point in each target grid according to the number of neighboring points and the moving speed of each position point in each target grid and the grid density of each target grid, including: performing normalization processing on the number of neighboring points and the moving speed of each position point in each target grid and the grid density of each target grid to obtain the normalized value of the number of neighboring points and the moving speed of each position point in each target grid and the normalized value of the grid density of each target grid; and multiplying the normalized value of the number of neighboring points and the moving speed of each position point in each target grid and the normalized value of the grid density of each target grid to obtain the clustering center weight of each position point in each target grid. The determining and judging module is configured to determine each user stay grid of the target user according to the clustering center weight of each position point in each target grid, and to judge the stay point of the target user according to the specific position information of each user stay grid. The determining and judging module is configured to determine each user stay grid of the target user according to the clustering center weight of each position point in each target grid, and to judge the stay point of the target user according to the specific position information of each user stay grid.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the user stay point judgment method in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium comprising a computer program, characterized in that, The computer program is executed by the processor to implement the user stay point judgment method in any one of claims 1 to 6.

Images