Itinerary residence point extraction method, system, electronic device and storage medium
By using the standard deviation verification algorithm to identify user travel trajectories, the problems of abnormal decomposition of dwell points and low recognition accuracy are solved, achieving efficient dwell point extraction and making it suitable for large-scale user trajectory processing.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA MOBILE GROUP ZHEJIANG
- Filing Date
- 2023-08-28
- Publication Date
- 2026-06-16
AI Technical Summary
Existing technologies suffer from problems such as abnormal splitting of user dwell points, low recognition accuracy, and high algorithm complexity when identifying user dwell points, and are particularly ineffective in processing large batches of user trajectories.
The standard deviation verification algorithm is used to identify user travel trajectories. The original location trajectory set and dwell point are determined by calculating the standard deviation of latitude and longitude. The target location trajectory set is extracted by combining the preset threshold and the dwell point identification process is optimized.
It improves the accuracy and robustness of dwell point identification, reduces algorithm complexity, and is suitable for processing large batches of user trajectories.
Smart Images
Figure CN117150156B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of computer technology, specifically to a method, system, electronic device, and storage medium for extracting travel stops. Background Technology
[0002] Current methods for identifying station locations in signaling data mainly include: 1. Using methods that merge consecutive identical base station trajectories to identify station locations. This method does not consider the possibility of base station jumps when the user's location remains unchanged, leading to abnormal station point splitting and inaccurate location. 2. Using complex unsupervised clustering algorithms such as K-means and DBSCAN, which only have a clustering effect on spatial data. Affected by the jumps and sparsity of signaling, these clustering algorithms perform poorly in joint time-space clustering, requiring auxiliary logic for secondary processing. Therefore, these algorithms are only suitable for clustering analysis of single or small numbers of user trajectories, and their high complexity makes them unsuitable for identifying station locations of hundreds of millions of user trajectories. 3. Using general machine learning algorithms to directly aggregate station locations without considering the characteristics of user signaling jumps in wireless communication technology. This method cannot specifically remove the impact of abnormal noise caused by signaling trajectory drift on station location identification, resulting in large station location identification deviations and reduced algorithm accuracy. Summary of the Invention
[0003] This application provides a method, system, electronic device, and storage medium for extracting travel stops, aiming to achieve lightweight identification while improving identification accuracy.
[0004] In a first aspect, embodiments of this application provide a method for extracting itinerary stops, including:
[0005] Arrange the user's trip trajectory data in chronological order to obtain the trip trajectory time series;
[0006] Based on the travel trajectory time series and standard deviation verification algorithm, the original location trajectory set is determined, and the original dwell point is determined based on the original location trajectory set;
[0007] Based on the original location trajectory set and the original dwell point, extract the first target location trajectory set.
[0008] In one embodiment, determining the original location trajectory set based on the travel trajectory time series and standard deviation verification algorithm, and determining the original dwell point based on the original location trajectory set, includes:
[0009] Obtain the occurrence time of the starting position point in the travel trajectory time series, and obtain all position point data to be processed based on the occurrence time and the preset dwell time threshold;
[0010] Calculate the standard deviation of latitude and longitude based on the latitude and longitude of all the location data to be processed;
[0011] The original location trajectory set is determined based on the latitude and longitude standard deviation;
[0012] The mean latitude and longitude is calculated based on the latitude and longitude of each location in the original location trajectory set, and the point corresponding to the mean latitude and longitude is determined as the original station point.
[0013] In one embodiment, extracting the first target location trajectory set based on the original location trajectory set and the original dwell points includes:
[0014] Determine the complete set of location trajectories within the designated area; the location points in the complete set of location trajectories are the location points arranged in chronological order.
[0015] Based on the original set of location trajectories and the complete set of location trajectories, a set of location points to be extracted is determined;
[0016] Based on the distance between the original dwell point and each location point in the set of location points to be extracted, extract the target location point set in the set of location points to be extracted;
[0017] The original set of location trajectories and the set of target location points are merged to obtain the first set of target location trajectories.
[0018] In one embodiment, extracting the target set of location points from the set of location points to be extracted based on the distance between the original dwell point and each location point in the set of location points to be extracted includes:
[0019] Starting from the first location point in the set of location points to be extracted, location points whose distance from the original dwell point is less than a preset threshold are identified as target location points. This process continues until an end location point is identified whose distance from the original dwell point is greater than or equal to the preset threshold. Furthermore, if all location points after the end location point are greater than or equal to the preset dwell time threshold and their distance from the original dwell point is greater than or equal to the preset threshold, then all target location points before the end location point are grouped together to obtain the target location point set.
[0020] In one embodiment, after extracting the first target location trajectory set based on the original location trajectory set and the original dwelling points, the method further includes:
[0021] Determine the dwell time of the last location point in the first target location trajectory set;
[0022] Based on the end time of the dwell point and the set of location points to be extracted, determine the set of remaining location trajectories;
[0023] Based on the set of remaining location trajectories, extract the set of remaining dwell points, and merge and optimize the set of remaining dwell points to obtain the set of second target location trajectories.
[0024] In one embodiment, merging and optimizing the set of remaining dwell points to obtain the second target location trajectory set includes:
[0025] Determine the distance between any two adjacent points in the set of remaining points;
[0026] If it is determined that the distance between any two adjacent dwelling points is less than a preset threshold, then the number of movement trajectory points between any two adjacent dwelling points is determined.
[0027] If the number of movement trajectory points between any two adjacent dwelling points is less than a preset number, then any two adjacent dwelling points are merged to obtain the second target location trajectory set.
[0028] In one embodiment, after extracting the first target location trajectory set based on the original location trajectory set and the original dwelling points, the method further includes:
[0029] Based on the first target location trajectory set, the second target location trajectory set, and the travel trajectory data, extract the user's mobile location set;
[0030] The first target location trajectory set, the second target location trajectory set, and the moving location set are integrated to obtain the stationary location information and the moving location information.
[0031] Secondly, embodiments of this application provide a system for extracting travel stops, comprising:
[0032] The data processing module is used to arrange the user's travel trajectory data in chronological order to obtain the travel trajectory time series;
[0033] The determination module is used to determine the original location trajectory set based on the travel trajectory time series and standard deviation verification algorithm, and to determine the original dwell point based on the original location trajectory set;
[0034] The dwell point extraction module is used to extract a first target location trajectory set based on the original location trajectory set and the original dwell points.
[0035] Thirdly, embodiments of this application provide an electronic device, which includes a memory, a processor, and a deterministic machine program stored in the memory and executable on the processor. When the processor executes the deterministic machine program, it implements the travel dwell point extraction method described in the first aspect.
[0036] Fourthly, embodiments of this application provide a non-transitory deterministic machine-readable storage medium, which includes a deterministic machine program. When the deterministic machine program is executed by a processor, it implements the process dwell point extraction method described in the first aspect.
[0037] Fifthly, embodiments of this application provide a computer product, which includes a deterministic program. When the deterministic program is executed by a processor, it implements the process dwell point extraction method described in the first aspect.
[0038] The travel stop point extraction method, system, electronic device, and storage medium provided in this application arrange user travel trajectory data in chronological order to obtain a travel trajectory time series; based on the travel trajectory time series and a standard deviation verification algorithm, an original location trajectory set is determined, and an original stop point is determined based on the original location trajectory set; based on the original location trajectory set and the original stop point, a first target location trajectory set is extracted. During the travel stop point extraction process, the standard deviation verification algorithm identifies user stop points and movement points. It demonstrates significant integration and recognition effects on user stop-state location trajectories with obvious jump characteristics in the travel trajectory data, exhibiting high robustness. Compared to machine learning clustering algorithms such as K-means and DBSCAN, it is more lightweight and improves recognition accuracy. Attached Figure Description
[0039] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0040] Figure 1 This is a flowchart of the itinerary stop point extraction method provided in the embodiments of this application;
[0041] Figure 2 This is a flowchart for determining the initial dwell point based on a sliding window;
[0042] Figure 3 This is a flowchart for extracting the first target's position trajectory set based on a sliding window;
[0043] Figure 4 This is a flowchart of the merging and optimization of the outposts;
[0044] Figure 5 This is a flowchart that outputs the stationary location information and the movement location information;
[0045] Figure 6 This is a structural diagram of the itinerary stop point extraction system provided in the embodiments of this application;
[0046] Figure 7 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation
[0047] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0048] like Figure 1 , Figure 1 This is a flowchart of a method for extracting itinerary stops provided in an embodiment of this application. An embodiment of this application provides a method for extracting itinerary stops, including:
[0049] Step 101: Arrange the user's travel trajectory data in chronological order to obtain the travel trajectory time series;
[0050] Step 102: Based on the travel trajectory time series and standard deviation verification algorithm, determine the original location trajectory set, and determine the original dwell point based on the original location trajectory set;
[0051] Step 103: Extract the first target location trajectory set based on the original location trajectory set and the original dwell point.
[0052] It should be noted that the itinerary stop point extraction method provided in this application embodiment is illustrated by taking the stop point extraction system as the execution subject.
[0053] Specifically, the stay point extraction system obtains the user's daily travel trajectory data, sorts the travel trajectory data in chronological order from front to back, and obtains the travel trajectory time series. The travel trajectory data is obtained based on operator data, which includes but is not limited to user signaling data, call detail records (CDRs), and MR data.
[0054] Furthermore, the dwell point extraction system employs a standard deviation verification algorithm to analyze the time series of the travel trajectory, identifying whether the user's location trajectory is in a dwell state, and obtaining the original location trajectory set. The standard deviation verification algorithm is used to verify the clustering characteristics of the user's location trajectory location points within a continuous time window; the denser the distribution of location points within the time window, the smaller the standard deviation of the latitude and longitude of the location data. Further, the dwell point extraction system determines the original dwell points based on the original location trajectory set.
[0055] Furthermore, the dwell point extraction system extracts the first target location trajectory set based on the original location trajectory set and the original dwell points.
[0056] The travel stop point extraction method provided in this application arranges the user's travel trajectory data in chronological order to obtain a travel trajectory time series; based on the travel trajectory time series and a standard deviation verification algorithm, it determines the original location trajectory set and the original stop points; based on the original location trajectory set and the original stop points, it extracts the first target location trajectory set. During the travel stop point extraction process, the standard deviation verification algorithm identifies user stop points and movement points. It demonstrates significant integration and recognition effects on user stop-state location trajectories with obvious jump characteristics in the travel trajectory data, exhibiting high robustness. Compared to machine learning clustering algorithms such as K-means and DBSCAN, it is more lightweight and improves recognition accuracy.
[0057] In one embodiment, determining the original location trajectory set based on the travel trajectory time series and standard deviation verification algorithm, and determining the original dwell point based on the original location trajectory set includes:
[0058] Obtain the occurrence time of the starting position point in the travel trajectory time series, and obtain all position point data to be processed based on the occurrence time and the preset dwell time threshold;
[0059] Calculate the standard deviation of latitude and longitude based on the latitude and longitude of all the location data to be processed;
[0060] The original location trajectory set is determined based on the latitude and longitude standard deviation;
[0061] The mean latitude and longitude is calculated based on the latitude and longitude of each location in the original location trajectory set, and the point corresponding to the mean latitude and longitude is determined as the original station point.
[0062] Specifically, step 1: The dwell point extraction system obtains the starting position point in the travel trajectory time series, that is, the first position point, as the starting position point p of the window. start Starting position point p start The corresponding occurrence time is t start .
[0063] Step 2: The dwell time extraction system uses a preset dwell time threshold as the minimum time threshold for dwell point identification, starting from the occurrence time t of the starting position point of the window. start The system acquires all location data and their corresponding latitude and longitude coordinates within a preset dwell time threshold window. The preset dwell time threshold is set according to actual conditions; in one embodiment, it is 30 minutes. Therefore, it can be understood that 30 minutes is the minimum time threshold for dwell point identification in the dwell point extraction system, starting from the occurrence time t of the initial location point within the window. start The system retrieves all location data to be processed and their corresponding latitude and longitude within the next 30-minute window. Further, the dwell point extraction system calculates the standard deviation σ of latitude and longitude based on the latitude and longitude of all the location data to be processed.
[0064] Step 3: The dwell point extraction system determines the original location trajectory set based on the latitude and longitude standard deviation σ. Specifically, if the latitude and longitude standard deviation σ is less than the preset standard deviation threshold σ0, the dwell point extraction system determines and generates a dwell point. The preset standard deviation threshold σ0 is set according to actual conditions. Therefore, it can be understood that the original location trajectory set of the dwell point is P′={p|t start <t p ≤t start +30min}. Further, the dwell point extraction system calculates the mean latitude and longitude of each location in the original location trajectory set, and determines the point corresponding to the mean latitude and longitude as the original dwell point p′. u Step 4: If the latitude and longitude standard deviation σ is determined to be greater than or equal to the preset standard deviation threshold σ0, the dwell point extraction system will slide the window and set the starting position point p of the current window. start The next location point is updated as the window start point, and steps 2 and 3 are executed again until the latitude and longitude standard deviation σ is less than the preset standard deviation threshold σ0 to determine the dwell point, or the window is slid to the last trajectory point to confirm that no dwell point has been generated. See details for further information. Figure 2 , Figure 2 This is a flowchart for determining the original dwell point based on a sliding window.
[0065] This application embodiment identifies user dwell points through the standard deviation verification algorithm. It has a significant effect on the integrated identification of user dwell state position trajectories with obvious jump characteristics in the travel trajectory data. It is highly robust and is more lightweight than machine learning clustering algorithms such as K-means and DBSCAN, while improving the identification accuracy.
[0066] In one embodiment, extracting a first target location trajectory set based on the original location trajectory set and the original dwell points includes:
[0067] Determine the complete set of location trajectories within the designated area; the location points in the complete set of location trajectories are the location points arranged in chronological order.
[0068] Based on the original set of location trajectories and the complete set of location trajectories, a set of location points to be extracted is determined;
[0069] Based on the distance between the original dwell point and each location point in the set of location points to be extracted, extract the target location point set in the set of location points to be extracted;
[0070] The original set of location trajectories and the set of target location points are merged to obtain the first set of target location trajectories.
[0071] It should be noted that the original residence point p′ obtained above u The final state of the non-stationary point, the original set of position trajectories P′ is only the minimum set of position trajectories confirming the existence of the stationary point. Therefore, let P be the set of all position trajectories within the stationary range, then... Let p be the last location point within the range of the dwelling point. end The last position point p end The corresponding time is t end The dwell point extraction system then determines the set of all location trajectories P as {p|t}. start <t p ≤t end andt end >t start +30min}, where the location points in the entire location trajectory set are the location points arranged in time sequence. Since operator location signaling inevitably contains noise, to avoid individual noisy locations affecting the dwell point identification results, the last location point p is set as... end The condition that must be met is the last position point p. end The previous position point p in the timing sequence end-1 Satisfy specific dis(p) end-1 ,p′ u )≤σ1, and the point set within the subsequent 30-minute window all satisfy the distance dis(p|t)≤σ1. end <t p ≤t end +30min,p′ u )>σ1, where σ1 is a preset threshold.
[0072] Therefore, the dwell point extraction system should satisfy the following condition to determine the set of points within the entire location trajectory set P:
[0073] P = {p|t} start <t p ≤t end ,and tend >t start +30min, and dis(p,p′) u )
[0074] ≤σ1,dis(p end-1 , p′ u )≤σ1,dis(p|t end <t p
[0075] ≤t end +30min,p′ u )>σ1}
[0076] Furthermore, the dwell point extraction system extracts the target location point set from the location point set based on the distance between the original dwell point and each location point in the location point set to be extracted.
[0077] Furthermore, the dwell point extraction system merges the original set of location trajectories and the set of target location points to obtain the first set of target location trajectories.
[0078] This application embodiment identifies user dwell points through the standard deviation verification algorithm. It has a significant effect on the integrated identification of user dwell state position trajectories with obvious jump characteristics in the travel trajectory data. It is highly robust and is more lightweight than machine learning clustering algorithms such as K-means and DBSCAN, while improving the identification accuracy.
[0079] In one embodiment, extracting a target set of location points from the set of location points to be extracted, based on the distance between the original dwell point and each location point in the set of location points to be extracted, includes:
[0080] Starting from the first location point in the set of location points to be extracted, location points whose distance from the original dwell point is less than a preset threshold are identified as target location points. This process continues until an end location point is identified whose distance from the original dwell point is greater than or equal to the preset threshold. Furthermore, if all location points after the end location point are greater than or equal to the preset dwell time threshold and their distance from the original dwell point is greater than or equal to the preset threshold, then all target location points before the end location point are grouped together to obtain the target location point set.
[0081] Specifically, the dwell point extraction system starts from the first location point in the set of location points to be extracted, and identifies location points whose distance from the original dwell point is less than a preset threshold as target location points, until an end location point is identified whose distance from the original dwell point is greater than or equal to the preset threshold. Furthermore, if all location points after the end location point have a distance from the original dwell point greater than or equal to the preset threshold, then all target location points before the end location point are aggregated to obtain the target location point set.
[0082] Let P be the set of original location trajectories. ′ The first position in the subsequent temporal sort is point p.
[0083] Step 1: Calculate the relationship between point p and the original dwell point p ′ u The distance dis(p,p) ′ u If the distance dis(p,p) is determined, ′ u If )≤σ1, where σ1 is a preset threshold, the dwell point extraction system will determine point p as the target location point and include it in the set of all location trajectories P, and continue to calculate the next location point and the original dwell point p. ′ u The distance is included in the set of all position trajectories P according to the conditions, until dis(p,p) appears. ′ u )>σ1 or point p reaches the last position point of the trajectory sequence.
[0084] Step 2: If it is determined that dis(p, p) ′ u If σ > 1, the dwell point extraction system then determines whether all points within a 30-minute window following point p satisfy dis(p, p) = σ1. ′ u If all points within a consecutive 30-minute window after determining point p satisfy dis(p, p) > σ1. ′ u If σ > 1, the dwell point extraction system will set the location point before point p as the last location point in the time sequence within the dwell point range, and its occurrence time will be the dwell point departure time t. pend If the conditions are not met, the dwell point extraction system determines point p as an abnormal jump point and discards it, then recalculates from the next location point according to step 1 until t is obtained. pend .
[0085] Step 3: Calculate the average latitude and longitude p of all points in the set P of all location trajectories. u As the center point of the dwelling point, therefore, the current location p of the dwelling point is obtained. uStart time t pstart and end time t pend The overall process is as follows Figure 3 , Figure 3 This is a flowchart for extracting the first target position trajectory set based on a sliding window.
[0086] This application embodiment identifies user dwell points through a standard deviation verification algorithm. It has a significant effect on the integrated identification of user dwell state position trajectories with obvious jump characteristics in the travel trajectory data. It is highly robust, lighter than machine learning clustering algorithms, and improves the identification accuracy.
[0087] In one embodiment, after extracting the first target location trajectory set based on the original location trajectory set and the original dwell points, the method further includes:
[0088] Determine the dwell time of the last location point in the first target location trajectory set;
[0089] Based on the end time of the dwell point and the set of location points to be extracted, determine the set of remaining location trajectories;
[0090] Based on the set of remaining location trajectories, extract the set of remaining dwell points, and merge and optimize the set of remaining dwell points to obtain the set of second target location trajectories.
[0091] Specifically, the dwell point extraction system determines the dwell point end time of the last location point in the first target location trajectory set based on the above steps, and determines the remaining location trajectory set based on the dwell point end time and the set of location points to be extracted.
[0092] Furthermore, the dwell point extraction system iterates through the remaining location trajectory set according to the above embodiment's process of determining the original dwell point based on a sliding window and extracting the first target location trajectory set based on a sliding window, extracting all remaining dwell points and the start and end times of all remaining dwell points, until the last trajectory point is processed.
[0093] Furthermore, the dwell point extraction system aggregates all remaining dwell points to obtain a set of remaining dwell points. It should be noted that, on the one hand, abnormal signaling jumps may cause dwell points to be incorrectly split, and on the other hand, users of dwell point data may require merging of consecutive dwell points over short distances. Therefore, it is necessary to merge and optimize the set of remaining dwell points to obtain a set of second target location trajectories.
[0094] In one embodiment, the remaining set of dwell points is merged and optimized to obtain a second target location trajectory set, including:
[0095] Determine the distance between any two adjacent points in the set of remaining points;
[0096] If it is determined that the distance between any two adjacent dwelling points is less than a preset threshold, then the number of movement trajectory points between any two adjacent dwelling points is determined.
[0097] If the number of movement trajectory points between any two adjacent dwelling points is less than a preset number, then any two adjacent dwelling points are merged to obtain the second target location trajectory set.
[0098] It should be noted that the merging principle is: 1. Close distance, that is, two consecutive dwelling points p u1 and p u2 1. The distance between them is less than a preset threshold; 2. Few behaviors, i.e., two consecutive dwell points p u1 and p u2 The number of movement trajectory points between them is less than the preset number.
[0099] Therefore, the dwell point extraction system merges and optimizes the remaining dwell point set based on the above two principles. Specifically, the dwell point extraction system determines the distance between any two adjacent dwell points in the remaining dwell point set. If the distance between any two adjacent dwell points is less than a preset threshold, the dwell point extraction system determines the number of movement trajectory points between any two adjacent dwell points, where the preset threshold is set according to actual conditions. If the number of movement trajectory points between any two adjacent dwell points is less than a preset number, the dwell point extraction system merges any two adjacent dwell points to obtain a second target location trajectory set, where the preset number is set according to actual conditions. This can also be further understood as:
[0100] Step 1: For each pair of consecutive dwell points in the remaining dwell point set, perform a merge rule check. If the condition is met, merge them. For the two consecutive dwell points p1 and p2 to be merged, the position of the merged dwell point is p1. u1 The start and end times are t respectively. p1start and t p2end .
[0101] Step 2: Repeat Step 1 to verify and merge the merged dwell point with the next dwell point.
[0102] Step 3: Repeat the process until all dwell points have been determined, then output the final set of dwell points, which is the second target location trajectory set. Refer to the detailed process below. Figure 4 The flowchart for optimizing the merging of residence points.
[0103] The embodiments of this application accurately reconstruct the complete trajectory dwell points of a single user.
[0104] In one embodiment, after extracting the first target location trajectory set based on the original location trajectory set and the original dwell points, the method further includes:
[0105] Based on the first target location trajectory set, the second target location trajectory set, and the travel trajectory data, extract the user's mobile location set;
[0106] The first target location trajectory set, the second target location trajectory set, and the moving location set are integrated to obtain the stationary location information and the moving location information.
[0107] Specifically, the stop point extraction system removes location points from the travel trajectory data that belong to the first target location trajectory set and the second target location trajectory set, extracting the user's mobile location set, i.e., the location data during non-stop periods, and tags it with mobile location information. Further, the stop point extraction system integrates the first target location trajectory set, the second target location trajectory set, and the mobile location set to obtain stop location information and mobile location information, as detailed in [reference needed]. Figure 5 , Figure 5 This is a flowchart that outputs the stationary location information and the movement location information.
[0108] The embodiments of this application use an iterative algorithm to accurately reconstruct the complete trajectory dwell points and movement points of a single user.
[0109] The itinerary stop point extraction system provided in the embodiments of this application is described below. The itinerary stop point extraction system described below can be referred to in correspondence with the itinerary stop point extraction method described above. Reference Figure 6 , Figure 6 This is a structural diagram of the itinerary stop point extraction system provided in this application embodiment. The itinerary stop point extraction system provided in this application embodiment includes:
[0110] Data processing module 601 is used to arrange the user's trip trajectory data in chronological order to obtain a trip trajectory time series;
[0111] The determination module 602 is used to determine the original location trajectory set based on the travel trajectory time series and standard deviation verification algorithm, and to determine the original dwell point based on the original location trajectory set;
[0112] The dwell point extraction module 603 is used to extract a first target location trajectory set based on the original location trajectory set and the original dwell points.
[0113] The travel stop point extraction system provided in this application arranges the user's travel trajectory data in chronological order to obtain a travel trajectory time series; based on the travel trajectory time series and a standard deviation verification algorithm, it determines the original location trajectory set and the original stop points based on the original location trajectory set; based on the original location trajectory set and the original stop points, it extracts the first target location trajectory set. During the travel stop point extraction process, the standard deviation verification algorithm identifies user stop points and movement points. It demonstrates significant integration and recognition effects on user stop-state location trajectories with obvious jump characteristics in the travel trajectory data, exhibiting high robustness. Compared to machine learning clustering algorithms such as K-means and DBSCAN, it is more lightweight and improves recognition accuracy.
[0114] In one embodiment, the determining module 602 is further configured to:
[0115] Obtain the occurrence time of the starting position point in the travel trajectory time series, and obtain all position point data to be processed based on the occurrence time and the preset dwell time threshold;
[0116] Calculate the standard deviation of latitude and longitude based on the latitude and longitude of all the location data to be processed;
[0117] The original location trajectory set is determined based on the latitude and longitude standard deviation;
[0118] The mean latitude and longitude is calculated based on the latitude and longitude of each location in the original location trajectory set, and the point corresponding to the mean latitude and longitude is determined as the original station point.
[0119] In one embodiment, the dwell point extraction module 603:
[0120] Determine the complete set of location trajectories within the designated area; the location points in the complete set of location trajectories are the location points arranged in chronological order.
[0121] Based on the original set of location trajectories and the complete set of location trajectories, a set of location points to be extracted is determined;
[0122] Based on the distance between the original dwell point and each location point in the set of location points to be extracted, extract the target location point set in the set of location points to be extracted;
[0123] The original set of location trajectories and the set of target location points are merged to obtain the first set of target location trajectories.
[0124] In one embodiment, the dwell point extraction module 603:
[0125] Starting from the first location point in the set of location points to be extracted, location points whose distance from the original dwell point is less than a preset threshold are identified as target location points. This process continues until an end location point is identified whose distance from the original dwell point is greater than or equal to the preset threshold. Furthermore, if all location points after the end location point are greater than or equal to the preset dwell time threshold and their distance from the original dwell point is greater than or equal to the preset threshold, then all target location points before the end location point are grouped together to obtain the target location point set.
[0126] In one embodiment, the itinerary stop point extraction system is also used for:
[0127] Determine the dwell time of the last location point in the first target location trajectory set;
[0128] Based on the end time of the dwell point and the set of location points to be extracted, determine the set of remaining location trajectories;
[0129] Based on the set of remaining location trajectories, extract the set of remaining dwell points, and merge and optimize the set of remaining dwell points to obtain the set of second target location trajectories.
[0130] In one embodiment, the itinerary stop point extraction system is also used for:
[0131] Determine the distance between any two adjacent points in the set of remaining points;
[0132] If it is determined that the distance between any two adjacent dwelling points is less than a preset threshold, then the number of movement trajectory points between any two adjacent dwelling points is determined.
[0133] If the number of movement trajectory points between any two adjacent dwelling points is less than a preset number, then any two adjacent dwelling points are merged to obtain the second target location trajectory set.
[0134] In one embodiment, the itinerary stop point extraction system is also used for:
[0135] Based on the first target location trajectory set, the second target location trajectory set, and the travel trajectory data, extract the user's mobile location set;
[0136] The first target location trajectory set, the second target location trajectory set, and the moving location set are integrated to obtain the stationary location information and the moving location information.
[0137] The specific embodiments of the itinerary stop point extraction system provided in this application are basically the same as the embodiments of the itinerary stop point extraction method, and will not be described in detail here.
[0138] Figure 7An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 7 As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730, and a communication bus 770, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other via the communication bus 770. The processor 710 can call the deterministic program in the memory 730 to execute the steps of the travel dwell point extraction method, such as including:
[0139] Arrange the user's trip trajectory data in chronological order to obtain the trip trajectory time series;
[0140] Based on the travel trajectory time series and standard deviation verification algorithm, the original location trajectory set is determined, and the original dwell point is determined based on the original location trajectory set;
[0141] Based on the original location trajectory set and the original dwell point, extract the first target location trajectory set.
[0142] Furthermore, the logical instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a deterministic machine-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This deterministic machine software product is stored in a storage medium and includes several instructions to cause a deterministic machine device (which may be a personal deterministic machine, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0143] On the other hand, embodiments of this application also provide a non-transient deterministic machine-readable storage medium, the non-transient deterministic machine-readable storage medium including a deterministic machine program, the deterministic machine program being stored on the non-transient deterministic machine-readable storage medium, and when the deterministic machine program is executed by a processor, the deterministic machine is able to perform the steps of the travel dwell point extraction method provided in the above embodiments, for example including:
[0144] Arrange the user's trip trajectory data in chronological order to obtain the trip trajectory time series;
[0145] Based on the travel trajectory time series and standard deviation verification algorithm, the original location trajectory set is determined, and the original dwell point is determined based on the original location trajectory set;
[0146] Based on the original location trajectory set and the original dwell point, extract the first target location trajectory set.
[0147] In another aspect, embodiments of this application also provide a computer product, the computer product including a determination machine program, the determination machine program being stored on the computer product, and when the determination machine program is executed by a processor, the determination machine is able to perform the steps of the process dwell point extraction method provided in the above embodiments, such as including:
[0148] Arrange the user's trip trajectory data in chronological order to obtain the trip trajectory time series;
[0149] Based on the travel trajectory time series and standard deviation verification algorithm, the original location trajectory set is determined, and the original dwell point is determined based on the original location trajectory set;
[0150] Based on the original location trajectory set and the original dwell point, extract the first target location trajectory set.
[0151] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.
[0152] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This deterministic machine software product can be stored in a deterministic machine-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a deterministic machine device (which may be a personal deterministic machine, a server, or a network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments.
[0153] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.
Claims
1. A method for extracting travel stops, characterized in that, include: Arrange the user's trip trajectory data in chronological order to obtain the trip trajectory time series; Based on the travel trajectory time series and standard deviation verification algorithm, the original location trajectory set is determined, and the original dwell point is determined based on the original location trajectory set; Based on the original location trajectory set and the original dwell point, a first target location trajectory set is extracted, including: Determine the complete set of location trajectories within the designated area; the location points in the complete set of location trajectories are the location points arranged in chronological order. Based on the original set of location trajectories and the complete set of location trajectories, a set of location points to be extracted is determined; Based on the distance between the original dwell point and each location point in the set of location points to be extracted, extract the target location point set in the set of location points to be extracted; The original set of location trajectories and the set of target location points are merged to obtain the first set of target location trajectories.
2. The method for extracting travel stops according to claim 1, characterized in that, The step of determining the original location trajectory set based on the travel trajectory time series and standard deviation verification algorithm, and determining the original dwell point based on the original location trajectory set, includes: Obtain the occurrence time of the starting position point in the travel trajectory time series, and obtain all position point data to be processed based on the occurrence time and the preset dwell time threshold; Calculate the standard deviation of latitude and longitude based on the latitude and longitude of all the location data to be processed; The original location trajectory set is determined based on the latitude and longitude standard deviation; The mean latitude and longitude is calculated based on the latitude and longitude of each location in the original location trajectory set, and the point corresponding to the mean latitude and longitude is determined as the original station point.
3. The method for extracting travel stops according to claim 2, characterized in that, The step of extracting the target set of location points from the set of location points to be extracted based on the distance between the original dwell point and each location point in the set of location points to be extracted includes: Starting from the first location point in the set of location points to be extracted, location points whose distance from the original dwell point is less than a preset threshold are identified as target location points. This process continues until an end location point is identified whose distance from the original dwell point is greater than or equal to the preset threshold. Furthermore, if all location points after the end location point are greater than or equal to the preset dwell time threshold and their distance from the original dwell point is greater than or equal to the preset threshold, then all target location points before the end location point are grouped together to obtain the target location point set.
4. The method for extracting travel stops according to claim 1, characterized in that, After extracting the first target location trajectory set based on the original location trajectory set and the original dwelling point, the process further includes: Determine the dwell time of the last location point in the first target location trajectory set; Based on the end time of the dwell point and the set of location points to be extracted, determine the set of remaining location trajectories; Based on the set of remaining location trajectories, extract the set of remaining dwell points, and merge and optimize the set of remaining dwell points to obtain the set of second target location trajectories.
5. The method for extracting travel stops according to claim 4, characterized in that, The process of merging and optimizing the remaining set of dwelling points to obtain the second target location trajectory set includes: Determine the distance between any two adjacent points in the set of remaining points; If it is determined that the distance between any two adjacent dwelling points is less than a preset threshold, then the number of movement trajectory points between any two adjacent dwelling points is determined. If the number of movement trajectory points between any two adjacent dwelling points is less than a preset number, then any two adjacent dwelling points are merged to obtain the second target location trajectory set.
6. The method for extracting travel stops according to claim 5, characterized in that, After extracting the first target location trajectory set based on the original location trajectory set and the original dwelling point, the process further includes: Based on the first target location trajectory set, the second target location trajectory set, and the travel trajectory data, extract the user's mobile location set; The first target location trajectory set, the second target location trajectory set, and the moving location set are integrated to obtain the stationary location information and the moving location information.
7. A system for extracting travel stops, characterized in that, The method for extracting travel stops as described in claim 1 includes: The data processing module is used to arrange the user's travel trajectory data in chronological order to obtain the travel trajectory time series; The determination module is used to determine the original location trajectory set based on the travel trajectory time series and standard deviation verification algorithm, and to determine the original dwell point based on the original location trajectory set; The dwell point extraction module is used to extract a first target location trajectory set based on the original location trajectory set and the original dwell points.
8. An electronic device comprising a memory, a processor, and a deterministic machine program stored in the memory and executable on the processor, characterized in that, When the processor executes the deterministic machine program, it implements the travel dwell point extraction method according to any one of claims 1 to 6.
9. A non-transitory deterministic machine-readable storage medium, the non-transitory deterministic machine-readable storage medium comprising a deterministic machine program, characterized in that, When the determination machine program is executed by the processor, it implements the travel dwell point extraction method according to any one of claims 1 to 6.