Signaling-based user travel residence point determination method and device, equipment and medium

Abstract

The invention provides a signaling-based user travel residence point determination method and device, equipment and a medium. The method comprises the following steps: sorting track points in travel track data of a user according to a time sequence to obtain a track point sequence, segmenting the track point sequence according to a preset first time interval to obtain multiple track point subsequences, and clustering the track point subsequences by adopting a density-based clustering algorithm to obtain multiple track point subsequences; and clustering the multiple sections of track point subsequences by adopting a time sliding window method to obtain a clustering set cluster, and finally determining the resident point of the user according to the clustering set cluster. According to the method provided by the invention, the track points moving back and forth in the same path are prevented from being clustered into the resident points, so that the identification states of the track points in the same path are correctly identified, and the identification accuracy of the resident points is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of communication technology, and in particular to a signaling-based user travel residence point determination method, device, equipment and medium. BACKGROUND

[0002] With the popularity of mobile communication devices such as smart phones, smart watches, and smart dash cams, operators have accumulated massive amounts of mobile location data resources, which can provide detailed and powerful data support for the construction of "smart transportation". When studying the travel characteristics of urban residents based on mobile location data, it is necessary to first identify the residence points in the user's trajectory. Residence point identification has wide applications in transportation, urban layout, public safety, and other fields.

[0003] In the prior art, methods for identifying residence points include distance-based identification, time-based identification, and density-based identification. Among them, the distance and time-based identification method identifies residence points by setting spatial distance thresholds and time span thresholds. The density-based identification method identifies residence points using point density, i.e., a large number of points appear in a region. By setting a density threshold, hot spot regions can be identified to determine residence points.

[0004] However, the existing density-based identification method can identify complete hot spot regions, but considering the randomness of actual user travel time and frequency, the diversity of travel methods, and the complexity of travel routes such as transfers and round trips, this residence point identification method may also have the problem of misjudging data points on the round trip path as residence points when the user's round trip path is the same. SUMMARY

[0005] The present application provides a signaling-based user travel residence point determination method, device, equipment and medium to solve the technical problem that electronic devices in the prior art may misjudge the identification state of data points when identifying residence points, resulting in clustering data points on the round trip path as residence points.

[0006] In one aspect, the present application provides a signaling-based user travel residence point determination method, comprising:

[0007] sorting the trajectory points in the user's travel trajectory data in chronological order to obtain a trajectory point sequence;

[0008] segmenting the trajectory point sequence according to a preset first time interval to obtain multiple trajectory point subsequences;

[0009] using a density-based clustering algorithm to cluster the multiple trajectory point subsequences using a time sliding window method to obtain a clustering collection cluster, the clustering collection cluster including a center positioning point of each clustering cluster obtained by clustering;

[0010] According to the cluster set, a stay point of the user is determined.

[0011] In a specific embodiment, before the trajectory points in the travel trajectory data of the user are sorted in time sequence to obtain a trajectory point sequence, the method further comprises:

[0012] Obtaining initial travel trajectory data of the user, the initial travel trajectory data comprising collected trajectory point data of the user, each trajectory point data comprising spatial position coordinates and collection time;

[0013] Performing denoising processing on the initial travel trajectory data to obtain denoised initial travel trajectory data;

[0014] Using equidistant interpolation points, data is supplemented to the denoised initial travel trajectory data to obtain time-continuous travel trajectory data.

[0015] In a specific embodiment, according to the cluster set, a stay point of the user is determined, comprising:

[0016] Based on time sequence, the positioning points in the cluster set are cut to obtain a plurality of positioning point clusters, wherein the cutting position of the positioning point cluster is the position where the time interval between adjacent positioning points is greater than a preset second time interval;

[0017] The first type of positioning point cluster in the plurality of positioning point clusters is deleted, the first positioning point and the last positioning point of the first type of positioning point cluster having a time difference less than a preset third time interval;

[0018] After the first type of positioning point cluster is deleted, the remaining plurality of positioning point clusters are sorted in time sequence, the time interval between the last positioning point of each positioning point cluster and the first positioning point of the next positioning point cluster is calculated, and if the time interval is less than a preset fourth time interval, the positioning point cluster and the next positioning point cluster are merged into one positioning point cluster to obtain a plurality of processed positioning point clusters; the fourth time interval is greater than the second time interval;

[0019] Positioning point clusters with a spatial distance less than a preset distance in the plurality of processed positioning point clusters are merged to obtain a plurality of merged positioning clusters; wherein the spatial distance between any two positioning point clusters is calculated according to the average value of the latitude and longitude of the positioning points in each positioning point cluster;

[0020] According to the positioning points in the plurality of merged positioning clusters, a stay point of the user is determined.

[0021] In an embodiment, the determining the residence point of the user according to the positioning points in the merged positioning clusters comprises:

[0022] The positioning points in the merged positioning clusters are determined as the residence point of the user, and the positioning points in the travel trajectory data that do not belong to the merged positioning clusters are determined as the moving points.

[0023] In an embodiment, the clustering the plurality of sub-sequence of trajectory points by using the density-based clustering algorithm and the time sliding window method comprises:

[0024] For the plurality of sub-sequence of trajectory points, the density-based clustering algorithm and the time sliding window method are used for clustering according to the preset radius parameter and the point density parameter, and a plurality of clustering clusters are obtained, each of which includes a plurality of positioning points.

[0025] The clustering cluster set is obtained according to the center positioning point in each clustering cluster.

[0026] In the clustering process, the time sliding window for each clustering includes at least two sub-sequences of trajectory points.

[0027] In an embodiment, the data supplementing the initial travel trajectory data after denoising by using the equidistant interpolation method to obtain the time-continuous travel trajectory data comprises:

[0028] For the missing time point in the initial travel trajectory data after denoising, the trajectory point data of the time point is determined according to the last trajectory point data before the time point and the first trajectory point data after the time point, and the time-continuous travel trajectory data is obtained.

[0029] In an embodiment, the trajectory data of the time point t is calculated by using the following formula:

[0030]

[0031] wherein, x represents the longitude of the time point t, y represents the latitude of the time point t, t represents the time point t a represents the time of the last trajectory point data before the time point t, t b represents the time of the first trajectory point data after the time point t, x a represents the longitude of the time point t a , y a represents the latitude of the time point t a , x b represents the longitude of the time point t b , y b represents the latitude of the time point tb latitude.

[0032] In a second aspect, the present application provides a signaling-based user travel residence point determination device, comprising:

[0033] A first processing module, configured to sort trajectory points in user travel trajectory data in time sequence to obtain a trajectory point sequence;

[0034] A second processing module, configured to segment the trajectory point sequence according to a preset first time interval to obtain a plurality of trajectory point subsequences;

[0035] A third processing module, configured to use a density-based clustering algorithm to cluster the plurality of trajectory point subsequences using a time sliding window method to obtain a clustering collection cluster, wherein the clustering collection cluster includes a center positioning point of each clustering cluster obtained by clustering;

[0036] A fourth processing module, configured to determine a residence point of the user according to the clustering collection cluster.

[0037] In a third aspect, the present application provides an electronic device, comprising:

[0038] A processor, and a memory connected with the processor in communication;

[0039] The memory stores computer execution instructions;

[0040] The processor executes the computer execution instructions stored in the memory to implement the signaling-based user travel residence point determination method of the first aspect.

[0041] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores computer execution instructions, and the computer execution instructions are executed by a processor to implement the signaling-based user travel residence point determination method of the first aspect.

[0042] The signaling-based user travel residence point determination method, device, equipment and medium provided by the present application first sort trajectory points in user travel trajectory data in time sequence to obtain a trajectory point sequence, then segment the trajectory point sequence according to a preset first time interval to obtain a plurality of trajectory point subsequences, then use a density-based clustering algorithm to cluster the plurality of trajectory point subsequences using a time sliding window method to finally obtain a clustering collection cluster including a center positioning point of each clustering cluster obtained by clustering, and finally determine a residence point of the user according to the clustering collection cluster. Such a technical means avoids clustering trajectory points under the same path for round trips as residence points, thereby realizing correct identification of the identification state of trajectory points under the same path and improving the accuracy of residence point identification. Attached Figure Description

[0043] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0044] Figure 1 A flowchart of an embodiment of a signaling-based method for determining user travel and stay points provided by this invention;

[0045] Figure 2 A flowchart of a second embodiment of a signaling-based user travel and stay point determination method provided by this invention;

[0046] Figure 3 A flowchart of another embodiment of the signaling-based user travel and stay point determination method provided by the present invention;

[0047] Figure 4 A flowchart of another method for determining user travel and stay points based on signaling, provided in embodiment four of the present invention;

[0048] Figure 5 A schematic diagram of a signaling-based user travel and stay point determination device provided in an embodiment of the present invention;

[0049] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0050] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0051] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0052] To clearly understand the technical solution of this application, the solutions of the prior art will be described in detail first.

[0053] Currently, the use of mobile communication devices such as smart phones, smart watches, and smart dash cams is increasing. Through continuous acquisition of signaling data, operators have accumulated massive mobile location data resources, which can provide detailed and powerful data support for the construction of "smart transportation". When studying the travel characteristics of urban residents based on mobile location data, it is first necessary to identify the dwell points in the user trajectory.

[0054] Currently, the method for identifying dwell points includes distance-based identification, time-based identification, and density-based identification. Distance-based identification identifies dwell points based on the distance between consecutive points in the user trajectory data. When the distance between two points is less than a certain threshold, it is considered that the user stays at that location. This method is simple and intuitive, but may not be accurate enough for identifying the shape of the dwell point. Time-based identification considers the length of time a user stays at a location. When a user stays at the same location for more than a certain value, it is identified as a dwell point. This method can better identify points that are stayed at for a long time. However, it is too subjective and difficult to judge. Density-based identification uses the density of points to identify dwell points, i.e., a large number of points appear in a region. By setting a density threshold, the complete hot spot region can be identified, thereby determining the dwell point. However, the back-and-forth movement along the same path may misjudge the moving point as a dwell point, which greatly reduces the accuracy of determining the user's travel dwell point based on signaling and reduces the reliability of the data.

[0055] First, the terms involved in the present application are explained:

[0056] 1. Signaling data

[0057] Signaling data refers to the data generated during the interaction between a mobile communication device and a wireless communication network. When a mobile communication device communicates externally, it interacts with nearby base stations, which record information such as the number of communication devices, communication time, and communication location and transmit it back to the data management center. After a long period of data accumulation, operators can obtain the continuous location trajectory of mobile communication devices. As a record of personal activity data, mobile phone signaling data contains a wealth of information, which provides a basis for mining user travel characteristics from signaling data.

[0058] 2. DBSCAN algorithm

[0059] DBSCAN (Density-Based Spatial Clustering of Applications with Noise, DBSCAN) algorithm is one of the most commonly used density clustering algorithms, which can identify clusters of arbitrary shape, and it divides high-density areas into a cluster and low-density areas as noise. By defining the field radius parameter Eps and the point density parameter Minpts, core points, boundary points and noise points are determined. Moreover, when clustering based on the density of DBSCAN algorithm, the number of clusters to be formed does not need to be known in advance.

[0060] Based on the description of the existing resident point identification scheme in the background art, it can be known that in the prior art, when the user makes a round trip in the same path, the electronic device often clusters the moving points at the same position as the resident point without considering the time factor, which leads to misjudgment of the determination of the user's travel resident point and reduces the recognition accuracy.

[0061] To solve the above technical problems, the technical concept of the present application is that the inventor found that when the user makes a round trip in the same path, the existing density-based identification method can identify the complete hot spot area, but cannot distinguish the moving points and resident points at the same position, and often clusters the trajectory points left by the user at the same position at different times as resident points, which greatly reduces the accuracy of resident point identification. Therefore, the inventor introduces the time sliding window method into the density-based clustering algorithm, thereby distinguishing the moving points and resident points at the same position, and improving the recognition accuracy.

[0062] The present application provides a user travel resident point determination method based on signaling, which is applied to various servers, computer devices, terminals and other electronic devices capable of processing data. The present application does not limit the specific device form.

[0063] Specifically, resident point identification has wide application in many fields such as transportation, city layout, public safety, etc. For example, in the field of transportation, by analyzing the resident points in the vehicle trajectory, the traffic flow can be optimized and the travel efficiency can be improved; in the field of city layout, the identification of resident points can help to find new interest places and provide reference for city layout and management.

[0064] The technical solutions of the present application and how the technical solutions of the present application solve the above technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of the present application will be described below with reference to the drawings.

[0065] The specific implementation scheme of the resident point determination method provided in the application will be introduced below.

[0066] Figure 1 A flowchart of a signaling-based user travel resident point determination method provided for the first embodiment of the application is shown in FIG. 1. Figure 1 As shown in the figure, the signaling-based user travel resident point determination method specifically includes the following steps:

[0067] S101: Sort the trajectory points in the user's travel trajectory data in time sequence to obtain a trajectory point sequence.

[0068] In order to identify the travel feature state of the user, it is necessary to first identify the resident point in the user's travel process. When a mobile communication device communicates externally, the nearby base station will record the device number, communication time and location data and transmit them back to the data management center. The electronic device can retrieve the required signaling data from the data management center, and use the user's travel trajectory data contained in the signaling data to identify the resident point in the user's travel process. Specifically, the travel trajectory data includes the user's multiple trajectory point data collected by the user, and each trajectory point data includes spatial position coordinates and collection time.

[0069] In this step, in order to distinguish the moving points and resident points on the same path, the electronic device arranges the trajectory points in time sequence before clustering the trajectory points in the user's travel trajectory data, thereby obtaining a trajectory point sequence, wherein the arrangement order can be from early to late in time or from late to early in time. For example, an employee passes through path A in the morning when going to work and passes through path A again in the afternoon when going home. When identifying the resident points of the employee throughout the day, it is necessary to distinguish the two times of passing through path A. Therefore, arranging the trajectory points in the employee's travel trajectory data throughout the day in time sequence from early to late can effectively avoid misjudging the round-trip moving points on the same path as resident points, thereby improving the identification accuracy.

[0070] S102: According to a preset first time interval, segment the trajectory point sequence to obtain multiple trajectory point subsequences.

[0071] In this step, in order to facilitate subsequent batch clustering analysis of the trajectory points, the trajectory sequence needs to be segmented into multiple trajectory point subsequences according to a preset first time interval before the clustering algorithm is performed. The segmented multiple trajectory point subsequences are t1, t2, t3, …, t nThe first time interval can be adaptively adjusted according to the length of the identified trajectory sequence and the window size of the time sliding window. For example, when identifying the stay points of a staff within one hour, the first time interval is set to two minutes. Then, the trajectory points are sorted in time sequence, and divided into 30 trajectory point subsequences, i.e., 0-2min, 2-4min, …, 58-60min.

[0072] In S103, a density-based clustering algorithm is used to cluster the multiple trajectory point subsequences by using a time sliding window method, to obtain a clustering collection cluster, wherein the clustering collection cluster includes the center positioning points of each clustering cluster obtained by clustering.

[0073] In this step, the multiple trajectory point subsequences are clustered by using a time sliding window method and a density-based clustering algorithm according to the preset radius parameter and point density parameter, to obtain multiple clustering clusters, each of which includes multiple center positioning points. Finally, the center positioning points in each clustering cluster are used to obtain a clustering collection cluster, wherein the time sliding window includes at least two trajectory point subsequences, and the window size of the sliding window can be adjusted according to the number of identified data points and the computing power of the electronic device.

[0074] Specifically, for the multiple trajectory point subsequences arranged in time sequence, the trajectory point subsequences in a first window are clustered from one end of the entire trajectory point sequence according to the preset radius parameter and point density parameter, to identify the point clusters in each travel chain trajectory that meet the clustering conditions. Then, the sliding window is moved to a second window, and the trajectory point subsequences in the second window are clustered by using the algorithm. The sliding window is continuously moved to the other end of the trajectory point sequence until all the trajectory point subsequences are clustered by using the algorithm. Finally, the center positioning points of each clustering cluster obtained by clustering are put into the clustering collection cluster.

[0075] In a possible implementation, the time sliding window can include three trajectory point subsequences. If the time sliding window includes three trajectory point subsequences, the electronic device first clusters the trajectory point subsequences t1, t2, and t3 by using the DBSCAN algorithm to identify the point clusters in the trajectory point subsequences that meet the clustering conditions. Then, the trajectory point subsequences t2, t3, and t4 are clustered by using the algorithm, and the trajectory point subsequences t3, t4, and t5 are clustered by using the algorithm. The iteration is performed in sequence until all the trajectory point subsequences are clustered. Finally, the center positioning points of each clustering cluster obtained by clustering are put into the clustering collection cluster, to facilitate the subsequent determination of the stay points.

[0076] In the specific application implementation of the scheme, it should be understood that the density-based clustering algorithm used can be any density-based clustering algorithm, including DBSCAN, OPTICS, HDBSCAN, DENCLUE, and Mean Shift algorithm, and the selection of the clustering algorithm is not limited in the present application.

[0077] S104: Determine the residence point of the user according to the clustering cluster.

[0078] In this step, the residence point of the user can be determined according to the center positioning points in the clustering cluster, and the residence point identifier obtained by the determination can be used for analyzing the travel characteristics of the user. Specifically, the center positioning points that are continuous in time and close in distance can be put into a cluster, and the center positioning points in the cluster are determined as the residence point of the user, and the remaining center positioning points are determined as the moving point of the user.

[0079] In the specific application implementation of the scheme, it should be understood that the center positioning points that are continuous in time and close in distance refer to that, after the center positioning points are sorted in time sequence, the time interval between the adjacent two center positioning points is less than the preset time interval, and at the same time, the spatial distance between the adjacent two clusters is less than the preset distance, wherein the preset time interval and spatial distance are determined according to the total time length or travel distance of the travel trajectory data studied, and the spatial distance between the adjacent two clusters is calculated according to the average value of the longitude and latitude of all center positioning points in each cluster.

[0080] In the user travel residence point determination method based on signaling provided by the embodiment, the trajectory points in the travel trajectory data of the user are sorted in time sequence to obtain a trajectory point sequence, then the trajectory point sequence is segmented into multiple trajectory point subsequences according to a preset first time interval, and then the multiple trajectory point subsequences are clustered and analyzed by using the time sliding window method combined with the density-based clustering algorithm, and then the clustering cluster that can be used to determine the residence point of the user is obtained. The embodiment fully considers the influence of the time factor on the residence point identification, avoids the electronic device from clustering the trajectory points of the back-and-forth movement on the same path as the residence point by combining the time sliding window with the density-based clustering algorithm, so as to realize the correct identification of the identification state of the trajectory points on the same path, and improve the accuracy of the residence point identification.

[0081] Figure 2 The flowchart of the second embodiment of the user travel residence point determination method based on signaling provided by the present application is shown in Figure 2 The scheme is mainly applied in the electronic device in the above-mentioned application scenario, and based on the above-mentioned first embodiment, before the trajectory points in the travel trajectory data of the user are sorted in time sequence to obtain a trajectory point sequence, the user travel residence point determination method based on signaling further includes the following steps:

[0082] S201: Obtain initial travel trajectory data of a user, the initial travel trajectory data including collected trajectory point data of the user, each trajectory point data including a spatial position coordinate and a collection time;

[0083] In the present solution, the foregoing embodiment is a detailed description of the clustering algorithm combined with the time sliding window, and the present embodiment introduces how the electronic device preprocesses the trajectory data after obtaining the initial travel trajectory data of the user.

[0084] In the present step, the electronic device can retrieve the required signaling data from the data management center, and identify the stay points in the travel process of the user by using the travel trajectory data of the user included in the signaling data, as described in the foregoing embodiment. When determining the all-day travel state of the user, the trajectory data obtained by the electronic device includes the all-day movement position trajectory points of the user, and each trajectory point data includes a spatial position coordinate and a collection time.

[0085] In the specific application implementation of the present solution, it should be understood that in real life, the user will be disturbed by external factors when the user has a business, therefore, the signaling data obtained by the electronic device will be mixed with a large amount of "noise", and therefore the travel trajectory data extracted from the signaling data also has noise. In addition, the travel trajectory data obtained by the electronic device does not contain the position information of the user at every moment, and there are some inaccurate data or no data at some moments.

[0086] S202: Perform denoising processing on the initial travel trajectory data to obtain denoised initial travel trajectory data;

[0087] In the present step, the obtained initial travel trajectory data needs to be denoised. Since in real life, the user will be disturbed by external factors when the user has a business, therefore, the initial travel trajectory data is mixed with a large amount of "noise", and the existence of the "noise" will affect the accuracy and reliability of the data, thereby causing great interference and errors to the stay point identification. Therefore, when identifying the stay points in the travel process of the user, the "noise" in the initial travel trajectory data needs to be removed by using a big data related denoising method.

[0088] In the specific implementation of the scheme, the large amount of "noise" mixed in the initial travel trajectory data includes repeated positioning, false switching information, ping-pong effect, memory effect, island effect, etc. For the "noise" in the initial travel trajectory data, denoising methods such as moving average method, wavelet transform denoising, Kalman filter, etc. can be used for elimination. The denoising methods of the initial travel trajectory data are various, and the specific selection of the method needs to be determined according to the characteristics of the data, the type of noise and the requirement of denoising. In actual application, it may be necessary to combine multiple denoising methods to achieve better denoising effect.

[0089] S203: The initial travel trajectory data after denoising is supplemented in a way of equidistant interpolation points to obtain time-continuous travel trajectory data.

[0090] In this step, due to the problems such as insufficient signal coverage, equipment failure, data processing error, etc. in real life, the user position information or communication events that should have been recorded are not correctly recorded or lost, therefore, the initial travel trajectory data after denoising may have missing trajectory point data, and direct identification will affect the accuracy. In the scheme, the sparse trajectory points are supplemented by the method of point supplementing to obtain time-continuous trajectory points, so that the individual travel and residence trajectory can be more intuitively observed, and it is also more conducive to the extraction of the residence point position.

[0091] Specifically, for the missing time point in the initial travel trajectory data after denoising, the trajectory point data of the time point is determined according to the last trajectory point data before the time point and the first trajectory point data after the time point, thereby the time-continuous travel trajectory data can be obtained.

[0092] In the scheme, it should be understood that whether there is a missing time point in the initial travel trajectory data after denoising is determined according to the preset positioning point frequency F, wherein the positioning point frequency F can be adjusted according to the computing power of the electronic device. If the time interval between two time-continuous trajectory points is greater than F seconds and there is no other trajectory point data between the two trajectory points, it is determined that there is a missing time point between the two trajectory points. In addition, the two trajectory points are determined as the last trajectory point data before the time point and the first trajectory point data after the time point, further, the time data and position data of the two trajectory points can be brought into the formula to calculate the trajectory point data of the missing time point, thereby the time-continuous travel trajectory data can be obtained.

[0093] Wherein, the trajectory data of the time point t is calculated by the following formula:

[0094]

[0095] wherein x represents the longitude of the time point t, y represents the latitude of the time point t, t a is the time of the last trajectory point data before the time point t, t b is the time of the first trajectory point data after the time point t, x a represents the longitude of the time point t a , y a represents the latitude of the time point t a , x b represents the longitude of the time point t b , y b represents the latitude of the time point t b .

[0096] In the specific implementation of the present solution, it should be understood that the number of interpolation points between the last trajectory point data before the time point and the first trajectory point data after the time point can be determined by the positioning point frequency F. Since the higher the interpolation point frequency, the greater the calculation amount of subsequent clustering, the positioning point frequency can be adjusted according to the computing power, and finally the longitude and latitude coordinate interpolation results of the positioning frequency of F seconds in the interval [t a , t b ] can be obtained.

[0097] In a possible implementation, the data after interpolation can be selected to become a continuous data set every second. The time axis is set to 0 at the zero point of a day, t a is the last trajectory point data before the time point t, t b is the first trajectory point data after the time point t, wherein t a = 3, t b = 6, and the calculation formula of the time point t is:

[0098] t = 3600 * hour + 60 * minute + second

[0099] Given that the coordinates of adjacent positioning points are (x a , y a , 3) and (x b , y b , 6) respectively, t = 4 and t = 5 need to be respectively brought into the following two-point straight line equation:

[0100]

[0101] After the above calculation, a continuous data set every second in the interval [3, 6] can be obtained.

[0102] The method for determining a user's travel residence point based on signaling provided in this embodiment includes the following steps: obtaining initial travel trajectory data of a user, performing denoising processing on the obtained initial travel trajectory data, and supplementing data of the denoised initial travel trajectory data by using an equidistant time interpolation method to obtain time-continuous travel trajectory data. The method fully considers that the initial travel trajectory data should be preprocessed before clustering the data points. The data analysis accuracy is improved by performing denoising processing on the initial travel trajectory data. The data integrity is ensured by supplementing data of the denoised initial travel trajectory data. The entire data preprocessing process facilitates the extraction of the residence point position and improves the accuracy of residence point identification.

[0103] Figure 3 A flowchart of a third embodiment of the method for determining a user's travel residence point based on signaling provided in the application is shown in FIG. 3, which mainly applies to the electronic device in the application scenario described above. The method can be combined with the first embodiment or the second embodiment. The residence point of the user is determined according to the clustering collection cluster, including the following steps. Figure 3

[0104] S301: cutting the positioning points in the clustering collection cluster based on time sequence to obtain a plurality of positioning point clusters, wherein the cutting position of the positioning point cluster is a position at which the time interval between adjacent positioning points is greater than a preset second time interval.

[0105] After the density-based clustering algorithm is used to cluster the multi-segment trajectory point subsequence by using the time sliding window method and the clustering collection cluster including the center positioning points of each clustering cluster is obtained, the clustering collection cluster can be optimized to make the finally obtained positioning point cluster satisfy the condition that the time distance between all positioning point clusters is greater than a fourth time interval and the spatial distance is greater than a preset distance, so that the clustering result is more intuitive and clear, facilitating the extraction of the residence point and improving the accuracy of residence point identification.

[0106] In this step, the positioning points that are continuous in time are grouped into one positioning point cluster, which is essentially to divide all the center positioning points in the clustering collection cluster into a plurality of positioning point clusters. Specifically, all the center positioning points in the clustering collection cluster are sorted in time sequence, and the clustering collection cluster is cut according to a preset second time interval. When the time interval between adjacent positioning points is greater than the preset second time interval, the clustering collection cluster is cut at this position, so that the two adjacent positioning points are in two different positioning point clusters. This operation is repeated until all the adjacent center positioning points in the clustering collection cluster are divided according to the requirement.

[0107] ​S302: deleting a first type of positioning point cluster in the plurality of positioning point clusters, wherein a time difference between a first positioning point and a last positioning point of the first type of positioning point cluster is less than a preset third time interval.

[0108] In this step, after the previous step, a plurality of positioning point clusters can be obtained. When the time difference between the first positioning point and the last positioning point in the positioning point cluster is less than the preset third time interval, it represents that the user stays at this place for a short time. Therefore, the positioning points in this positioning cluster cannot be identified as a stay point. Therefore, the first type of positioning point cluster should be deleted.

[0109] S303: sorting the plurality of positioning point clusters remaining after the first type of positioning point cluster is deleted in time order, calculating a time interval between the last positioning point of each positioning point cluster and the first positioning point of the next positioning point cluster, and if the interval is less than a preset fourth time interval, merging the positioning point cluster and the next positioning point cluster into one positioning point cluster to obtain a plurality of processed positioning point clusters, wherein the fourth time interval is greater than the second time interval.

[0110] In this step, in order to make the final positioning point cluster meet the requirement that the time distance between all positioning point clusters is greater than the fourth time interval, the plurality of positioning point clusters remaining after the first type of positioning point cluster is deleted should be sorted in time order. The time interval between the last positioning point of each positioning point cluster and the first positioning point of the next positioning point cluster is calculated. If the time interval between the two positioning points is less than the fourth time interval, the two positioning point clusters are merged. The merging process can make the division between positioning point clusters more clear and explicit, and can improve the accuracy of identifying stay points in the subsequent process.

[0111] In the specific application of this scheme, it should be understood that the fourth time interval should be greater than the second time interval. Because step S301 is to preliminarily divide the center positioning point in the clustering set cluster into a plurality of positioning point clusters, this step is to further optimize the positioning point cluster based on step S301. The time interval between the plurality of positioning point clusters can be larger, so that the electronic device can more quickly and accurately identify the stay point in the subsequent process.

[0112] S304: merging the positioning point clusters with a spatial distance less than a preset distance in the plurality of processed positioning point clusters to obtain a plurality of merged positioning clusters; wherein the spatial distance between any two positioning point clusters is calculated according to the average value of the longitude and latitude of the positioning points in each positioning point cluster.

[0113] In this step, in order to make the final obtained positioning point cluster meet the condition that the spatial distance between all positioning point clusters is greater than the preset distance, the remaining multiple positioning point clusters after the above deletion and merging processing are sorted in time sequence, the longitude and latitude average of all positioning points in each positioning point cluster is calculated, and if the spatial distance between two adjacent positioning point clusters is less than the preset distance, the two positioning point clusters are merged. The merging processing can make the division between the positioning point clusters more clear and explicit, and can improve the accuracy of subsequent residence point identification.

[0114] In the specific application implementation of the present solution, it should be understood that the spatial distance between two adjacent positioning point clusters is determined according to the longitude and latitude average of all positioning points in each positioning point cluster. The position coordinates of each positioning point cluster are calculated by the longitude and latitude average of the positioning points in the positioning point cluster, so that the spatial distance between two adjacent positioning point clusters can be obtained by subtracting the position coordinates of the two adjacent positioning point clusters.

[0115] S305: determining the residence point of the user according to the positioning points in the merged multiple positioning clusters.

[0116] In this step, after obtaining the multiple positioning point clusters after merging processing, the electronic device can determine the residence point of the user according to the positioning points in the positioning point cluster as above. Specifically, the positioning points in the merged multiple positioning clusters are determined as the residence point of the user, and the positioning points in the travel trajectory data that do not belong to the merged multiple positioning clusters are determined as the moving points. According to the determination of the residence point of the user in the travel process, the travel characteristics of the user can be further analyzed.

[0117] In the user travel residence point determination method based on signaling provided by the present embodiment, after obtaining the clustering collection cluster, the positioning points in the clustering collection cluster are cut based on time sequence to initially obtain multiple positioning point clusters, the positioning point clusters with short time intervals are deleted, then the remaining multiple positioning point clusters are sorted in time sequence, the positioning point clusters are merged according to the preset fourth time interval and spatial distance, and finally, the residence point of the user is determined according to the positioning points in the merged multiple positioning clusters. After obtaining the clustering collection cluster, the present embodiment performs clustering post-processing on the obtained collection cluster, and through deletion and merging processing, the time distance between the finally obtained positioning point clusters is greater than the fourth time interval, and the spatial distance is greater than the preset distance, which further limits the range to which the residence point belongs, so that the clustering result is more intuitive and explicit, and the accuracy of residence point identification is improved.

[0118] Based on the residence point determination method provided in any of the above embodiments, the following takes a server provided by a terminal device for an operator and trajectory points as all-day trajectory data of a single user as an example to illustrate the present scheme through a specific example. Figure 4 The flowchart of the fourth embodiment of the residence point determination method for user travel based on signaling provided by the present application.

[0119] Specifically, in the present embodiment, the server can determine the residence points in the all-day travel process of the user according to the obtained all-day trajectory data of the user. The residence point determination in the present scheme specifically includes the following steps:

[0120] S1: The server retrieves the all-day trajectory data of the target user from the data storage center.

[0121] S2: The server uses a denoising method such as the moving average method, the wavelet transform method, etc. to eliminate the "noise" in the data. The specific method to be selected depends on the characteristics of the data, the type of noise, and the denoising requirements.

[0122] S3: The server uses the equidistant interpolation method to supplement the initial travel trajectory data after denoising to obtain time-continuous travel trajectory data.

[0123] S4: The server sets the parameter conditions of the clustering algorithm according to the actual clustering requirements, including the domain radius parameter and the point density parameter.

[0124] S5: The server sorts the trajectory points in the travel trajectory data of the user in time sequence to obtain a trajectory point sequence, and divides the trajectory point sequence according to a preset first time interval to obtain multiple trajectory point subsequences. Then, the time sliding window method is used to perform the DBSCAN clustering algorithm on the multiple trajectory point subsequences to obtain a clustering collection.

[0125] S6: The server performs clustering post-processing on the obtained clustering collection. First, the server cuts the positioning points in the clustering collection based on time sequence to obtain multiple positioning point clusters. Then, the positioning point cluster whose time difference between the first positioning point and the last positioning point is less than a preset third time interval is deleted. Then, the remaining multiple positioning point clusters are sorted in time sequence, and the time interval between the last positioning point of each positioning point cluster and the first positioning point of the next positioning point cluster is calculated. If the time interval is less than a preset fourth time interval, the positioning point cluster and the next positioning point cluster are merged into one positioning point cluster to obtain the processed multiple positioning point clusters. Finally, the mean of the longitude and latitude of the positioning points in each positioning point cluster is calculated, and the positioning point clusters with a spatial distance less than a preset distance in the processed multiple positioning point clusters are merged.

[0126] S7: The server determines the user's all-day travel state according to the processed multiple positioning point clusters. The positioning points in the multiple positioning clusters are determined as the user's stay points, and the positioning points in the travel trajectory data that do not belong to the multiple positioning clusters after merging are determined as the moving points.

[0127] In the scheme, the server first performs the preprocessing operations of denoising and point supplementing on the initial travel trajectory data, which improves the accuracy of data analysis and ensures the integrity of data, and provides convenience for the extraction of the stay point positions. Then, the server uses the time sliding window to perform the density-based clustering algorithm, which avoids clustering the trajectory points in the same path that move back and forth as stay points, so as to correctly identify the state of the trajectory points in the same path. In addition, the server performs post-processing on the obtained clustering collection clusters, and through a series of deletion and merging processing, the clustering result is more intuitive and clear. The determination process of the stay point in the scheme improves the accuracy of stay point identification.

[0128] Figure 5 A signaling-based user travel stay point determination device structure diagram is provided for the embodiments of the application. The signaling-based user travel stay point determination device can be used to implement the technical solutions in any of the preceding method embodiments, as shown in Figure 5 The signaling-based user travel stay point determination device 40 includes:

[0129] A first processing module 401 is configured to sort the trajectory points in the user's travel trajectory data in chronological order to obtain a trajectory point sequence.

[0130] A second processing module 402 is configured to segment the trajectory point sequence according to a preset first time interval to obtain multiple trajectory point subsequences.

[0131] A third processing module 403 is configured to use a density-based clustering algorithm to cluster the multiple trajectory point subsequences using a time sliding window method to obtain a clustering collection cluster, and the clustering collection cluster includes the center positioning points of each clustering cluster obtained by clustering.

[0132] A fourth processing module 404 is configured to determine the user's stay points according to the clustering collection cluster.

[0133] The signaling-based user travel stay point determination device provided in the embodiment is used to execute the technical solutions in the preceding method embodiments, and the implementation manner and technical effects are similar, which will not be described here.

[0134] On the basis of the above-mentioned embodiment of the signaling-based user travel residence point determination apparatus 40, in a specific implementation, before the trajectory points in the user's travel trajectory data are sorted in time sequence to obtain a trajectory point sequence, the signaling-based user travel residence point determination apparatus 40 can further include:

[0135] The fifth processing module 405 is configured to acquire initial travel trajectory data of a user, the initial travel trajectory data including a plurality of trajectory point data of the user collected, each trajectory point data including a spatial position coordinate and a collection time;

[0136] The sixth processing module 406 is configured to perform denoising processing on the initial travel trajectory data to obtain denoised initial travel trajectory data.

[0137] The seventh processing module 407 is configured to perform data supplementing on the denoised initial trajectory point data in an equidistant time insertion manner to obtain time-continuous travel trajectory data.

[0138] The signaling-based user travel residence point determination apparatus provided in the embodiment is used to execute the technical solutions in the foregoing method embodiments, has similar implementation principles and technical effects, improves the data analysis precision by performing denoising processing on the initial travel trajectory data, ensures the data integrity by performing data supplementing on the denoised initial travel trajectory data, provides convenience for the extraction of the residence point position in the subsequent data preprocessing process, and improves the residence point recognition accuracy.

[0139] In a specific implementation, to determine the residence point of the user according to the cluster set, the fourth processing module 404 is specifically configured to:

[0140] cut the positioning points in the cluster set in time sequence to obtain a plurality of positioning point clusters, wherein the cutting position of the positioning point cluster is a position at which the time interval between adjacent positioning points is greater than a preset second time interval;

[0141] delete a first type of positioning point cluster in the plurality of positioning point clusters, the first type of positioning point cluster having a time difference between a first positioning point and a last positioning point that is less than a preset third time interval;

[0142] sort the plurality of positioning point clusters remaining after the first type of positioning point cluster is deleted in time sequence, calculate a time interval between the last positioning point of each positioning point cluster and a first positioning point of a next positioning point cluster, and if the time interval is less than a preset fourth time interval, merge the positioning point cluster and the next positioning point cluster into one positioning point cluster to obtain a plurality of processed positioning point clusters; the fourth time interval is greater than the second time interval.

[0143] merge the positioning point clusters with a spatial distance less than the preset distance in the processed plurality of positioning point clusters to obtain a plurality of merged positioning clusters; wherein the spatial distance between any two positioning point clusters is calculated according to the average longitude and latitude of the positioning points in each positioning point cluster;

[0144] determine the stay point of the user according to the positioning points in the plurality of merged positioning clusters.

[0145] On the basis of the above-mentioned embodiments, the fourth processing module 404 is further specifically configured to:

[0146] determine the positioning points in the plurality of merged positioning clusters as the stay point of the user, and determine the positioning points not belonging to the plurality of merged positioning clusters in the travel trajectory data as the moving point.

[0147] On the basis of the above-mentioned embodiments, in one specific implementation of the present solution, the third processing module 403 is specifically configured to:

[0148] cluster the plurality of trajectory point subsequences according to the preset radius parameter and point density parameter by using a time sliding window method and a density-based clustering algorithm to obtain a plurality of clustering clusters, each of which includes a plurality of positioning points;

[0149] obtain the clustering cluster according to the center positioning point in each clustering cluster;

[0150] wherein, in the clustering process, the time sliding window for each clustering includes at least two trajectory point subsequences.

[0151] Optionally, the seventh processing module 407 is specifically configured to:

[0152] determine the trajectory point data of the missing time point in the denoised initial travel trajectory data according to the last trajectory point data before the time point and the first trajectory point data after the time point to obtain the time-continuous travel trajectory data.

[0153] Optionally, the trajectory data of the time point t is calculated by using the following formula:

[0154]

[0155] wherein, x represents the longitude of the time point t, y represents the latitude of the time point t, time point t a is the time of the last trajectory point data before the time point t, time point t b is the time of the first trajectory point data after the time point t, x a represents the longitude of the time point t a , and y aRepresenting time point t a latitude, x b Representing time point t b longitude, y b Representing time point t b Latitude.

[0156] The signaling-based user travel and stay point determination device provided in any of the foregoing embodiments is used to execute the technical solutions in the foregoing method embodiments. The implementation methods and technical effects are similar, and will not be described again here.

[0157] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Figure 6 As shown, the electronic device 50 includes:

[0158] The electronic device 50 may include a processor 501 with one or more processing cores, a memory 502 with one or more computer-readable storage media, a communication component 503, and other components. The processor 501, memory 502, and communication component 503 are connected via a bus 504.

[0159] In the specific implementation process, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to execute the above-described signaling-based user travel and stay point determination method.

[0160] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0161] In the above Figure 6 In the illustrated embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0162] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0163] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, the bus in the drawings of the present application does not limit to only one bus or one type of bus.

[0164] In some embodiments, a computer program product is also provided, which includes a computer program or instructions, which, when executed by a processor, implement the steps of any of the above signaling-based user travel residence point determination methods.

[0165] The specific implementation of each operation can refer to the foregoing embodiments, which will not be described here.

[0166] Those skilled in the art can understand that all or part of the steps of the various methods of the above embodiments can be completed by instructions, or by relevant hardware controlled by the instructions, which can be stored in a computer readable storage medium and loaded and executed by a processor.

[0167] The embodiments of the present application provide a computer readable storage medium, which stores a plurality of instructions, which can be loaded by a processor to execute the steps of any of the signaling-based user travel residence point determination methods provided by the embodiments of the present application.

[0168] The storage medium can include a Read Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and the like.

[0169] The present application also provides a computer program product or a computer program, which includes computer instructions stored in a computer readable storage medium.

[0170] Since the instructions stored in the storage medium can execute the steps of any of the signaling-based user travel residence point determination methods provided by the embodiments of the present application, the beneficial effects of any of the signaling-based user travel residence point determination methods provided by the embodiments of the present application can be achieved, which will be described in detail in the foregoing embodiments, and will not be described here.

[0171] Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the application being indicated by the following claims.

[0172] It is to be understood that the application is not limited to the precise construction herein disclosed and shown in the drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the application. The scope of the application is limited only by the claims that follow.

Claims

1. A method for determining user travel and stay points based on signaling, characterized in that, include: The trajectory points in the user's travel trajectory data are sorted in chronological order to obtain a trajectory point sequence; According to a preset first time interval, the trajectory point sequence is divided into multiple trajectory point sub-sequences; A density-based clustering algorithm is used to cluster the multiple trajectory point subsequences using a time sliding window method to obtain cluster sets, wherein each cluster set includes the center location point of each cluster obtained by clustering. Based on the cluster set, the user's residence point is determined.

2. The method according to claim 1, characterized in that, Before sorting the trajectory points in the user's travel trajectory data according to time order to obtain the trajectory point sequence, the method further includes: The user's initial travel trajectory data is obtained, which includes multiple trajectory point data collected from the user. Each trajectory point data includes spatial location coordinates and collection time. The initial travel trajectory data is denoised to obtain denoised initial travel trajectory data; By using equal-time interpolation, the denoised initial travel trajectory data is supplemented to obtain time-continuous travel trajectory data.

3. The method according to claim 1 or 2, characterized in that, Determining the user's dwell point based on the cluster set includes: Based on the time sequence, the location points in the cluster are cut into multiple location point clusters. The cutting position of the location point cluster is the position where the time interval between adjacent location points is greater than a preset second time interval. The first type of positioning point cluster is deleted from the plurality of positioning point clusters, and the time difference between the first positioning point and the last positioning point of the first type of positioning point cluster is less than the preset third time interval. After deleting the first type of location point cluster, the remaining multiple location point clusters are sorted in chronological order. The time interval between the last location point of each location point cluster and the first location point of the next location point cluster is calculated. If the time interval is less than a preset fourth time interval, the location point cluster and the next location point cluster are merged into one location point cluster to obtain multiple processed location point clusters. The fourth time interval is greater than the second time interval. The spatial distance between multiple location point clusters that are less than a preset distance after processing is merged to obtain multiple merged location clusters; wherein, the spatial distance between any two location point clusters is calculated based on the average latitude and longitude of the location points in each location point cluster. The user's dwell point is determined based on the location points in the merged multiple location clusters.

4. The method according to claim 3, characterized in that, Determining the user's dwell point based on the location points in the merged multiple location clusters includes: The location points in the merged multiple location clusters are determined as the user's dwell points, and the location points in the travel trajectory data that do not belong to the merged multiple location clusters are determined as movement points.

5. The method according to claim 1 or 2, characterized in that, The density-based clustering algorithm is used to cluster the multiple trajectory point sequences using a time sliding window method to obtain cluster sets, including: For the multiple trajectory point sequences, clustering is performed using a time sliding window method and a density-based clustering algorithm based on preset radius parameters and point density parameters to obtain multiple clusters, each of which includes multiple positioning points; The cluster set is obtained based on the central location point in each cluster; In the clustering process, each clustering time window includes at least two sequences of trajectory points.

6. The method according to claim 2, characterized in that, The method of using equal-time interpolation to supplement the denoised initial travel trajectory data to obtain time-continuous travel trajectory data includes: For the missing time points in the denoised initial travel trajectory data, the trajectory point data of the time point is determined based on the last trajectory point data before the time point and the first trajectory point data after the time point, thus obtaining the time-continuous travel trajectory data.

7. The method according to claim 6, characterized in that, The trajectory data at time point t is calculated using the following formula: Where x represents the longitude of time point t, y represents the latitude of time point t, and time point t a The time of the last trajectory point data before time point t, where time point t is the time. b x represents the time of the first trajectory point data after time point t. a Representing time point t a longitude, y a Representing time point t a latitude, x b Representing time point t b longitude, y b Representing time point t b Latitude.

8. A signaling-based user travel and stay point determination device, characterized in that, include: The first processing module is used to sort the trajectory points in the user's travel trajectory data according to the time order to obtain the trajectory point sequence; The second processing module is used to divide the trajectory point sequence according to a preset first time interval to obtain multiple trajectory point sub-sequences. The third processing module is used to use a density-based clustering algorithm to cluster the multiple trajectory point subsequences using a time sliding window method to obtain cluster sets, wherein the cluster sets include the center positioning point of each cluster obtained by clustering; The fourth processing module is used to determine the user's dwell point based on the cluster set.

9. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 7.

Images