CN113742607B - Stay position recommending method based on geographical track of principal - Google Patents
Stay position recommending method based on geographical track of principal Download PDFInfo
- Publication number
- CN113742607B CN113742607B CN202010481631.5A CN202010481631A CN113742607B CN 113742607 B CN113742607 B CN 113742607B CN 202010481631 A CN202010481631 A CN 202010481631A CN 113742607 B CN113742607 B CN 113742607B
- Authority
- CN
- China
- Prior art keywords
- track
- stay
- smooth
- domain function
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000004364 calculation method Methods 0.000 claims abstract description 5
- 238000012545 processing Methods 0.000 claims description 10
- 230000000694 effects Effects 0.000 claims description 9
- 239000013598 vector Substances 0.000 claims description 9
- 239000003795 chemical substances by application Substances 0.000 claims description 8
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 abstract description 7
- 230000007547 defect Effects 0.000 abstract description 3
- 230000001419 dependent effect Effects 0.000 abstract description 3
- 230000008859 change Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 150000002148 esters Chemical class 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Position Fixing By Use Of Radio Waves (AREA)
Abstract
The invention provides a stay position recommending method based on a geographical track of a principal, and relates to the field of track analysis. The stay position recommending method based on the geographical track of the principal comprises a smooth track dividing algorithm and a stay domain function. The method is applied to the track person searching model by using the track stay domain function method to primarily identify the approximate area of the frequent stay position, overcomes the defects of overlarge data quantity and low accuracy of the traditional track analysis, solves the problem of too many dependent parameters of the track searching of the model in the prior art, divides the track of the principal into a plurality of smooth tracks by the smooth track dividing algorithm so as to reduce the later workload, improve the efficiency of the model, and has very fast algorithm and effectively improve the identification efficiency of the stay points by using the stay domain function and the calculation involved in the mark grading and the link.
Description
Technical Field
The invention relates to the technical field of track analysis, in particular to a stay position recommending method based on a geographical track of a principal.
Background
When a court executive bureau performs a court officer to find out a belief-losing executive person, the situation that the executive person frequently occurs to hide the trace occurs, and the current executive officer can only pass through the originally registered identity card address or the original contact information by searching for the executive person; because the complex location data cannot be effectively analyzed, once each executed person does not work or live at the original address, searching for the executed person becomes very troublesome.
The current traffic travel track analysis method is mainly focused on the DBSCAN proposed by Ester and the like for a density-based clustering algorithm, two parameters of an Eps neighborhood and a MinPts are set, and the connected search clusters are continuously expanded outwards in a classification mode with the density being reachable based on sample points. Birandom et al define spatial distance by using Eps1, and define non-spatial distance by using Eps2, and propose ST-DBSCAN algorithm of space-time clustering. Tran et al change Eps in DBSCAN into time linear neighborhood, minPts into minimum time length, and propose TrigDBSCASN algorithm. Gong et al introduces a direction change coefficient and an abnormal point proportion in the track, and proposes a C-DBSCAN algorithm aiming at time sequence limitation and direction change limitation of parking points. G.Agamennoni, J.Nieto, E.Nebot proposes an algorithm to find the salient locations in an automatic way, using two types of places, namely a low speed area and a loading station. The dwell index is proposed to Long Gang and the like, and the space-time aggregation degree of the track is reflected in the number of the neighborhood inner points, the space distance and the dwell time. He Yuanhao and the like combine DBSCAN with geographic semantics, extract clusters through density clustering, and then combine the semantics according to the reverse geographic coding of the cluster center. Zhou Yang, yang Chao propose a new space-time clustering algorithm AT-DBSCAN to identify parking points in a track, and can provide a certain basis for identifying travel modes and travel purposes. Yuan Hua, qian Yu, yang Rui have studied the problem of efficiently mining the location of a user's point of interest in a given spatial area and frequent travel sequences, i.e., frequent paths, from a large amount of GPS trajectory data. Wearing dew passes through trip test design, has gathered the complete trip chain GPS data of individual to utilize three kind trip endpoint identification algorithm respectively: the method has the advantages that the travel endpoints can be identified and evaluated based on a rule identification algorithm, a density-based spatial clustering algorithm DBSCAN and a density-based spatial-temporal clustering algorithm ST-DBScan, and the travel endpoints of most travel types can be identified, but the technology has the defects of overlarge analysis data volume and low accuracy, and meanwhile has the problem that model track searching dependent parameters are too many, so that the method is urgent for a court executive office to find out a believed executive.
Disclosure of Invention
(one) solving the technical problems
The invention aims to overcome the defects and the shortcomings of the prior art, and provides a stay position recommending method based on a geographical track of a principal, which is applied to a track person searching model by using a track stay domain function method to primarily identify the approximate area of frequent stay positions, overcomes the shortcomings of overlarge data quantity and low accuracy of the traditional track analysis, and solves the problem of too many dependent parameters of track searching of the model in the prior art.
(II) technical scheme
In order to achieve the above purpose, the invention is realized by the following technical scheme: the stay position recommending method based on the geographical track of the principal comprises a smooth track dividing algorithm and a stay domain function, and the method comprises the following steps:
s1, preprocessing the original data, including data cleaning and filling missing data by a linear interpolation method;
s2, dividing the track into a plurality of smooth tracks through a smooth track dividing algorithm and processing the smooth tracks by using a stay domain function, wherein the result of the smooth track dividing can ensure that characteristic points in the track, namely fixed active points and frequent paths of a principal can be effectively dug under the condition of less information loss;
s3, processing each smooth track by using a stay domain function, marking important places as geographical areas which are significant to users through scoring of the tracks and calculation related to links, deducing activities through data, and associating the activities with different places.
S4, gathering and classifying similar points according to a time threshold value of a certain position, constructing a place hierarchy through a time scale and a position scale, and calculating and identifying travel endpoints and frequent stay positions
S5, overlapping the position trace on the map, wherein the frequently accessed place is regarded as an important place. Furthermore an off-the-shelf GPS unit is connected to a person and processes the collected data to infer him or his activity, as well as an important series of places.
Preferably, the application step of the dwell domain function is as follows:
s1, firstly, processing data into two ordered setsAnd->
S2, the first group comprises position vectorsAnd is referred to as tracking, and the second group containsFeature vector
S3, summarizing any additional information about the agent at the same time point, wherein the agent only comprises the instant speed V when extracting the low-speed area n As a feature;
s4, selecting a corresponding position set with the score of 1Wherein, segmentation:the set contains all the locations of the current agent belonging to the low speed region.
Preferably, the residence domain function formula (V is velocity) is:
preferably, the feature vector may consist of raw data obtained directly from GPS readings, such as velocity and head values, as well as other values derived from the data.
Preferably, the smooth track dividing algorithm is as follows: given xi i For the ith GPS point of a certain individual, a series of GPS points with time sequence represent the individual's slave xi i Move to xi n Is a track TP of (2); tp= { ζ 1 ξ 2 …ξ iξi+1 …ξ n -a }; suppose that xi i (i=k+1, … n) inThe mapping on is ζ' i Then xi i And xi n The Euclidean distance between the two points is point xi i Is provided with a disturbance of the position of (a),by studying the sub-track STP= { ζ k ξ k+1 …ξ n Dots in } surrounding ∈>Position disturbance in directionDetecting whether a sub-track is a smooth track or not; giving a sub-trackAnd a predetermined small disturbance threshold τ 0 > 0, if satisfied->We consider that the position perturbation variation is not very large for all points within the STP (at the same time τ k+1 The position disturbance of (a) varies greatly), the sub-track being a smooth track, denoted SST (ζ) i ,ξ j ). Wherein xi j Is a turning point between two smooth tracks, namely a characteristic point; at this time, the track division problem can be embodied as finding a set of feature points ζ j So as to satisfy the two formulas.
(III) beneficial effects
The invention provides a stay position recommending method based on a geographical track of a principal. The device comprises the following
The beneficial effects are that:
1. the method starts with the analysis of the traffic travel track of the executed person, provides the potential geographic position information of the executed person for the executive officer, and improves the case execution efficiency.
2. The invention can divide the principal track into a plurality of smooth sub-tracks by using a smooth track dividing algorithm so as to reduce the workload in the later period and improve the efficiency of the model.
3. The invention uses the stay domain function, the algorithm is very fast through the mark scoring and the calculation related in the link, and the efficiency of identifying the stay point is effectively improved.
Drawings
FIG. 1 is a schematic flow chart of the present invention;
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Examples:
as shown in fig. 1, an embodiment of the present invention provides a stay location recommendation method based on a geographical track of a principal, including a smooth track dividing algorithm and a stay domain function, which is characterized in that: the method comprises the following steps:
s1, preprocessing the original data, including data cleaning and filling missing data by a linear interpolation method;
s2, dividing the track into a plurality of smooth tracks through a smooth track dividing algorithm and processing the smooth tracks by using a stay domain function so as to reduce the workload of the later stage, improve the efficiency of the model, and ensure that the characteristic points in the track, namely the fixed active points and the frequent paths of the principal, can be effectively dug under the condition of less information loss by the aid of the result of smooth track dividing;
s3, processing each smooth track by using a stay domain function, marking important places in the smooth tracks as geographical areas which are significant to users, such as residence addresses, workplaces, houses of friends and shopping centers, deducing activities by data and associating the activities with different places through scoring the tracks, wherein the related calculation is very simple, the algorithm is very fast, and the efficiency of identifying the stay points can be improved.
S4, gathering and classifying similar points according to a time threshold value of a certain position, constructing a place hierarchy through a time scale and a position scale, and calculating and identifying travel endpoints and frequent stay positions
S5, overlapping the position trace on the map, wherein the frequently accessed place is regarded as an important place. Furthermore an off-the-shelf GPS unit is connected to a person and processes the collected data to infer him or his activity, as well as an important series of places.
The dwell domain function formula (V is velocity) is:
the feature vector may consist of raw data obtained directly from GPS readings, such as velocity and head values, and other values derived from the data, the smooth trajectory partitioning algorithm being: given xi i For the ith GPS point of a certain individual, a series of GPS points with time sequence represent the individual's slave xi i Move to xi n Is a track TP of (2); tp= { ζ 1 ξ 2 …ξ iξi+1 …ξ n -a }; suppose that xi i (i=k+1, … n) inThe mapping on is ζ' i Then xi i And xi n The Euclidean distance between the two points is point xi i Is provided with a disturbance of the position of (a),by studying the sub-track STP= { ζ k ξ k+1 …ξ n Dots in } surrounding ∈>Position disturbance in direction Detecting whether a sub-track is a smooth track or not; giving a sub-track->And a predetermined small disturbance threshold τ 0 > 0, if satisfied->We consider that the position perturbation variation is not very large for all points within the STP (at the same time τ k+1 The position disturbance of (a) varies greatly), the sub-track being a smooth track, denoted SST (ζ) i ,ξ j ). Wherein xi j Is a turning point between two smooth tracks, namely a characteristic point; at this time, the track division problem can be embodied as finding a set of feature points ζ j In order to satisfy the above two formulas, through dividing the orbit into sub-trajectories first to carry out preliminary screening to the orbit, reduce the work load of follow-up model, improved the operating efficiency of model, stay domain function can be used in the orbit and seek people model to discern the approximate region of frequent stay position, its application step is:
s1, firstly, processing data into two ordered setsAnd->
S2, the first group comprises position vectorsAnd is referred to as tracking, and the second group contains feature vectors
S3, summarizing any additional agents at the same time pointInformation including only the instantaneous velocity V when extracting the low velocity region n As a feature;
s4, selecting a corresponding position set with the score of 1The set contains all the locations of the current agent belonging to the low speed region.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (2)
1. The stay position recommending method based on the geographical track of the principal comprises a smooth track dividing algorithm and a stay domain function, and is characterized in that: the method comprises the following steps:
s1, preprocessing the original data, including data cleaning and filling missing data by a linear interpolation method;
s2, dividing the track into a plurality of smooth tracks through a smooth track dividing algorithm and processing the smooth tracks by using a dwell domain function;
s3, processing each smooth track by using a stay domain function, marking important places as geographical areas significant to users through scoring of the tracks and calculation related to links, deducing activities through data, and associating the activities with different places;
s4, gathering and classifying similar points according to a time threshold value of a certain position, constructing a place hierarchy through a time scale and a position scale, and calculating and identifying travel endpoints and frequent stay positions
S5, superposing the position trace on the map, wherein the frequently accessed place is regarded as an important place; furthermore, connecting an off-the-shelf GPS unit to a person and processing the collected data to infer his or her activity, as well as a significant series of places;
the application steps of the residence domain function are as follows:
s1, firstly, processing data into two ordered setsAnd->
S2, the first group comprises position vectorsAnd is referred to as tracking, and the second group contains feature vectors
S3, summarizing any additional information about the agent at the same time point, wherein the agent only comprises the instant speed V when extracting the low-speed area n As a feature;
s4, selecting a corresponding position set with the score of 1The set contains all the locations of the current agent belonging to the low speed region;
the dwell domain function formula is:
wherein V is the speed;
the smooth track dividing algorithm is as follows: given xi i For the ith GPS point of a certain individual, a series of GPS points with time sequence represent the individual's slave xi i Move to xi n Is a track TP of (2); tp= { ζ 1 ξ 2 …ξ i ξ i+1 …ξ n -a }; suppose that xi i (i=k+1, … n) inThe mapping on is ζ' i Then xi i And xi n The Euclidean distance between the two points is point xi i Is disturbed by the position of->By studying the sub-track STP= { ζ k ξ k+1 …ξ n Dots in } surrounding ∈>Position disturbance in direction ∈>Detecting whether a sub-track is a smooth track or not; giving a sub-track->And a predetermined disturbance threshold tau 0 > 0, if satisfied->Then we consider that the position perturbation variation is not very large for all points within the STP, at the same time +.>The sub-track is a smooth track, denoted SST (ζ) i ,ξ j ) The method comprises the steps of carrying out a first treatment on the surface of the Wherein xi j Is a turning point between two smooth tracks, namely a characteristic point; at this time, the track division problem can be embodied as finding a set of feature points ζ j So as to satisfy the two formulas.
2. The stay location recommendation method based on a geographical trajectory of a party according to claim 1, wherein: the feature vector may consist of raw data obtained directly from the GPS readings, as well as other values derived from the data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010481631.5A CN113742607B (en) | 2020-05-28 | 2020-05-28 | Stay position recommending method based on geographical track of principal |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010481631.5A CN113742607B (en) | 2020-05-28 | 2020-05-28 | Stay position recommending method based on geographical track of principal |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113742607A CN113742607A (en) | 2021-12-03 |
CN113742607B true CN113742607B (en) | 2023-12-08 |
Family
ID=78727905
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010481631.5A Active CN113742607B (en) | 2020-05-28 | 2020-05-28 | Stay position recommending method based on geographical track of principal |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113742607B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104850604A (en) * | 2015-05-04 | 2015-08-19 | 华中科技大学 | Tensor-based user track mining method |
CN107589435A (en) * | 2017-09-05 | 2018-01-16 | 成都新橙北斗智联有限公司 | A kind of Big Dipper GPS track stops analysis method |
CN107657572A (en) * | 2017-09-13 | 2018-02-02 | 北京城建设计发展集团股份有限公司 | Dwell point recognition methods and system based on the equidistant space-time trajectory data of high frequency |
CN108242145A (en) * | 2016-12-26 | 2018-07-03 | 高德软件有限公司 | Abnormal track point detecting method and device |
CN109947136A (en) * | 2019-02-28 | 2019-06-28 | 清华大学 | A kind of collaboration actively perceive method towards unmanned plane group target searching |
CN109948701A (en) * | 2019-03-19 | 2019-06-28 | 太原科技大学 | A kind of data clustering method based on space-time relationship between track |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8020100B2 (en) * | 2006-12-22 | 2011-09-13 | Apple Inc. | Fast creation of video segments |
US20140287389A1 (en) * | 2013-03-14 | 2014-09-25 | The Regents Of The University Of California | Systems and methods for real-time adaptive therapy and rehabilitation |
-
2020
- 2020-05-28 CN CN202010481631.5A patent/CN113742607B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104850604A (en) * | 2015-05-04 | 2015-08-19 | 华中科技大学 | Tensor-based user track mining method |
CN108242145A (en) * | 2016-12-26 | 2018-07-03 | 高德软件有限公司 | Abnormal track point detecting method and device |
CN107589435A (en) * | 2017-09-05 | 2018-01-16 | 成都新橙北斗智联有限公司 | A kind of Big Dipper GPS track stops analysis method |
CN107657572A (en) * | 2017-09-13 | 2018-02-02 | 北京城建设计发展集团股份有限公司 | Dwell point recognition methods and system based on the equidistant space-time trajectory data of high frequency |
CN109947136A (en) * | 2019-02-28 | 2019-06-28 | 清华大学 | A kind of collaboration actively perceive method towards unmanned plane group target searching |
CN109948701A (en) * | 2019-03-19 | 2019-06-28 | 太原科技大学 | A kind of data clustering method based on space-time relationship between track |
Non-Patent Citations (3)
Title |
---|
Double-DQN based path smoothing and tracking control method for robotic vehicle navigation;Wenyu Zhang 等;《Computers and Electronics in Agriculture》;第166卷;1-11 * |
Spatial-temporal grid clustering method based on frequent stay point recognition;Bin Zhang 等;《Neural Computer and Applications》;9247-9255 * |
基于速度的轨迹停留点识别算法;蔡小路 等;《计算机系统应用》;第29卷(第4期);214-219 * |
Also Published As
Publication number | Publication date |
---|---|
CN113742607A (en) | 2021-12-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106462627B (en) | Analyzing semantic places and related data from multiple location data reports | |
Wang et al. | Anomalous trajectory detection and classification based on difference and intersection set distance | |
Buchin et al. | Similarity of trajectories taking into account geographic context | |
Cao et al. | Automatic user identification method across heterogeneous mobility data sources | |
Yu et al. | Trajectory outlier detection approach based on common slices sub-sequence | |
CN105701123A (en) | Passenger-vehicle relationship identification method and apparatus | |
CN104850604A (en) | Tensor-based user track mining method | |
CN108153867A (en) | User trajectory Forecasting Methodology and device based on temporal regularity | |
Li et al. | Robust inferences of travel paths from GPS trajectories | |
CN109190656B (en) | Indoor semantic track marking and complementing method under low-sampling positioning environment | |
Li et al. | A trajectory restoration algorithm for low-sampling-rate floating car data and complex urban road networks | |
Nishida et al. | Extracting arbitrary-shaped stay regions from geospatial trajectories with outliers and missing points | |
El Hafyani et al. | Learning the micro-environment from rich trajectories in the context of mobile crowd sensing: Application to air quality monitoring | |
Tiwari et al. | Mining popular places in a geo-spatial region based on GPS data using semantic information | |
CN106294485B (en) | Determine the method and device in significant place | |
CN113611115A (en) | Vehicle track clustering method based on road network sensitive characteristics | |
CN113742607B (en) | Stay position recommending method based on geographical track of principal | |
Li et al. | TORD problem and its solution based on big trajectories data | |
Alowayr et al. | A review of trajectory data mining applications | |
CN110555174B (en) | R-tree-based track path recommendation method | |
Liu et al. | Mining method based on semantic trajectory frequent pattern | |
Chen et al. | Traffic travel pattern recognition based on sparse Global Positioning System trajectory data | |
Qiu et al. | GPS trace mining for discovering behaviour patterns | |
Verstockt et al. | Collaborative Bike Sensing for Automatic Geographic Enrichment: Geoannotation of road\/terrain type by multimodal bike sensing | |
Li et al. | A GPS Trajectory Segmentation Method for Transportation Mode Inference |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |