CN112215666A - Characteristic identification method for different trip activities based on mobile phone positioning data - Google Patents
Characteristic identification method for different trip activities based on mobile phone positioning data Download PDFInfo
- Publication number
- CN112215666A CN112215666A CN202011211769.XA CN202011211769A CN112215666A CN 112215666 A CN112215666 A CN 112215666A CN 202011211769 A CN202011211769 A CN 202011211769A CN 112215666 A CN112215666 A CN 112215666A
- Authority
- CN
- China
- Prior art keywords
- time
- point
- space
- trip
- travel
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/029—Location-based management or tracking services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W8/00—Network data management
- H04W8/18—Processing of user or subscriber data, e.g. subscribed services, user preferences or user profiles; Transfer of user or subscriber data
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- Signal Processing (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a characteristic identification method for different travel activities based on mobile phone positioning data, which forms travel chains with different travel activities and time-space correlation characteristics with time-space correlation characteristics by performing time-space kernel cluster analysis and time-space correlation characteristic analysis on OD residence points of the mobile phone positioning data, performs verification and algorithm optimization on an obtained time-space residence point set, analyzes groups with different travel activities and characteristics, and finally forms an OD matrix travel table based on a traffic cell, thereby providing a big data support meeting the characteristics of the time-space travel activity chains for traffic planning and traffic demand management. According to the method, the advantages of spatial activity correlation analysis and the respective advantages of the OD matrix generation table are combined, the stationing point sequence with the characteristics of the space-time activity target is finally formed, the rationality and the accuracy of the OD matrix of different traffic trip targets are further improved, and the cost is effectively reduced.
Description
Technical Field
The invention relates to the technical field of traffic planning and traffic demand management, in particular to a characteristic identification method for different trip activities based on mobile phone positioning data, which can provide big data support for meeting the characteristics of different trip activities for traffic planning and traffic demand management.
Background
The traditional traffic trip survey is a processing process of statistical analysis based on individual traffic trip survey data, surveys such as trip purpose, trip mode, trip time, trip distance and the like need to be sampled and carried out through individuals of different crowds, and analysis processing results can reflect urban traffic demands and space-time distribution characteristics.
The analysis and processing process based on the traditional transportation survey needs to consume a large amount of manpower, material resources, capital and time, and cannot be frequently carried out. With the popularization of mobile phone terminals, the method for acquiring the user traffic travel information through the mobile phone positioning technology has the advantages of low cost, wide coverage range and the like, so that mobile phone data is used as an important supplement of the existing traffic data acquisition technology, and good technical support can be provided for the extraction of OD (origin-destination) characteristics and different travel target characteristics of residents during time and space travel.
At present, the characteristic judgment of the trip activity aim aiming at the mobile phone positioning technology is mainly based on the clustering technology or the POI identification technology, and then the judgment of the trip activity aim is carried out on the clustering points. When OD analysis is performed, the travel purpose judgment lacks effective integration of time-space correlation OD travel, and a one-day travel activity chain cannot be effectively generated. Therefore, the technology for judging the travel activity purpose based on the clustering technology or the POI identification technology ignores the overall characteristics of a complete travel activity chain in one day, and causes travel activity misjudgment which does not meet the characteristics of the travel OD activity purpose.
Disclosure of Invention
Aiming at the existing OD travel stationing point generation technology and combining the clustering judgment advantages of the occupational characteristics and the living activity travel activity characteristics, the invention provides a low-cost and high-accuracy characteristic identification method for different travel activities based on mobile phone positioning data, which comprises the following specific steps:
step 1: acquiring mobile phone signaling data and internet surfing data, wherein the mobile phone signaling data comprises user portrait label attribute data;
step 2: and (3) extracting mobile phone trigger data for three months by using the mobile phone signaling data and the internet surfing data collected in the step (1) to judge the position: firstly, carrying out spatial clustering on places with three months constant occurrence, carrying out high-frequency analysis according to the occurrence frequency of a stay time interval at the spatial clustering points, and carrying out occupational judgment on the high-frequency points meeting the time and space requirements, wherein the residence judgment is to be combined with the night constant on-off frequency to carry out frequency weight analysis judgment, the employment judgment is to be combined with a user age label to carry out frequent item set resident employment characteristic judgment based on duration, the residence and employment of a user are obtained, namely the occupation distribution characteristic is obtained, and then the residence distribution characteristic is matched with a time sequence to generate the time sequence with the occupation characteristic;
and step 3: based on the time sequence with the occupation characteristics in the step 2, determining OD (origin-destination) residence points of the spatio-temporal dimensions, namely, based on the time, distance and speed of vehicle displacement, finding out residence points meeting spatio-temporal displacement conditions through a heuristic search algorithm, matching the residence points with occupation data of a user, performing spatio-temporal kernel clustering analysis based on spatio-temporal association on a point set with central clustering characteristics and spatio-temporal trip characteristics, marking the residence points meeting occupation distribution characteristics, further performing spatio-temporal clustering analysis on the residence points not meeting occupation distribution characteristics, further determining whether the residence points belong to the residence points, and forming an OD set sequence with the spatio-temporal association occupation characteristics;
and 4, step 4: based on the OD set sequence with the time-space associated job and live characteristics in the step 3, judging the live activity trip of the time-space dimension, extracting all resident points of the user's home trip for live activity time-space trip analysis, performing time-space frequent item set analysis of the live activity trip, marking the resident points meeting the live activity trip characteristics to generate a live activity trip OD set, and marking the resident points not meeting the job distribution and the live activity trip characteristics as other resident points to generate an OD set sequence sharing other trip purposes;
in the step 4, the judgment of the liveness trip is specifically as follows: performing space kernel clustering analysis and judgment on the bioactivity based on a time sequence with the occupation characteristics, performing space clustering on a time space OD resident point of a certain user when a user goes out of a base house, and reforming a space clustering point cluster; analyzing frequent item sets with different residence time lengths based on the spatial clustering point clusters, and designing weights according to the principle that the longer the residence time is, the more the occurrence times are, and the more frequent the interaction with the family is; then, according to the weight values of the frequent item sets, counting the probability of the weight liveness destination of the spatial clustering point cluster, and then calculating the maximum value of the probability of the weight liveness destination, wherein the maximum value of the probability of the weight in the spatial clustering point cluster is the liveness destination;
in the step 4, the OD trip judgment of the liveness trip activity objective having the time-space correlation characteristic specifically includes: based on an OD (origin-destination) stay point set with a temporal-spatial correlation characteristic and an identification of a liveness destination, in a spatial dimension, whether the distance of a non-working stay point set meets a spatial distance threshold of the liveness destination or not is judged, meanwhile, in a temporal-spatial dimension, a time sequence is judged, whether a certain stay time length is met or not and liveness travel time distribution is met, if the temporal-spatial correlation characteristic of a life travel is met, a one-time liveness destination travel is formed, the travel is identified as a life travel for the current travel, and other stay point data which do not meet the temporal-spatial correlation characteristic are identified as other travels;
and 5: performing further time-space correlation analysis on the time-space OD set sequences of different trip activities obtained in the step (4), and further judging whether the time space has correlation characteristics or not by combining with the functional area identification of land utilization, and performing further time-space kernel clustering to form a time-space stationary point set with the time-space trip purposes;
step 6: dotting mobile phone signaling data of different trip characteristic crowds on a map to perform trip characteristic classification, completing verification of trip characteristics of the characteristic crowds, verifying the time-space stationing point set obtained in the step 5, judging whether the algorithm meets characteristics of different trip activities, if so, verifying to be qualified, otherwise, further optimizing the time-space trip characteristic analysis algorithm in the steps 3 and 4, and performing trip characteristic analysis on special crowds;
and 7: according to the traffic cell division scheme, the operator base station is matched with the traffic cell, the origin-destination point of each traveler is mapped to the corresponding traffic cell, and finally an OD matrix travel table based on different travel activities of the traffic cell is formed.
Preferably, the mobile phone trigger data comprises a user number, a timestamp, a longitude, a latitude, an individual age, a gender, whether students or retirees.
Preferably, in the step 2, the residence judgment specifically includes: firstly, carrying out time sequence sequencing on the same mobile phone number on extracted mobile phone trigger data within three months, and defining the time range of residence judgment as 21:00 in the evening to 07:00 in the morning; then carrying out spatial clustering on the mobile phone data with the same user number to form a spatial clustering point cluster; analyzing frequent item sets with different stay durations based on the spatial clustering point clusters, performing weight analysis by combining with the power-on and power-off event occurrence places of users at night, and performing weight design according to the principle that the longer the stay time is, the more the occurrence times are, and the more the power-on and power-off event occurrence places are, the larger the weight is; and then, according to the weight values of the frequent item sets, counting the weight residence probability of the spatial clustering point clusters, and calculating the maximum value of the weight residence probability in the spatial clustering point clusters, wherein the maximum value of the weight probability in the spatial clustering point clusters is the residence.
Preferably, in the step 2, the employment judgment specifically is: firstly, carrying out time series sequencing on the same mobile phone number on extracted mobile phone trigger data in half a year, defining the time range of employment place judgment as 9:00 in the daytime to 16:00 in the afternoon, eliminating retired old people and minors according to the attribute data of the user portrait label of an operator, and then judging the employment places of the rest of people; carrying out spatial clustering on the mobile phone data with the same user number to form a spatial clustering point cluster; analyzing frequent item sets with different stay durations based on the spatial clustering point clusters, and carrying out weight design according to the principle that the longer the stay time is, the more the occurrence times are, the larger the weight is, and if the maximum probability of the weighted employment place is greater than a membership threshold value and the corresponding spatial clustering point cluster and the residence place are not in the same cluster, judging the occupation place; otherwise, the position is not judged as the employment place position.
Preferably, the heuristic search algorithm in step 3 is specifically: for track point data sorted by time of a certain user, for one of the stationing points S _ i, the distance D _ i _ i +1 from the i +1 th point is calculated, if the distance D _ i _ i +2 from the i +2 th point to the stationing point S _ i > D _ i _ i +1, the maximum distance maxDist is D _ i _ i +2, and so on, the farthest distance point is calculated to be D _ i _ i + n, if the distance from the stationing point S _ i to the n +1 th point is greater than the distance from the stationing point S _ i to the n +2 th point, and the distance from the stationing point S _ i to the n +1 th point is greater than the distance threshold, and the speed from the stationing point S _ i to the n +2 th point is less than the speed threshold, the stationing state is determined.
Preferably, the method for determining the spatio-temporal core clustering residence point with occupational association characteristics in step 3 is as follows: based on an OD (origin-destination) stay point set with a spatio-temporal correlation characteristic, firstly, performing clustering analysis on a spatio-temporal dimension, and if distances from a user spatio-temporal stay point M1 or a user spatio-temporal stay point M2 to a place-of-employment clustering center are smaller than a distance threshold d1, considering that the user spatio-temporal stay point M1 and the user spatio-temporal stay point M2 are likely to form a place-of-employment stay and a trip; recursion to the 3 rd point M3, space dimension matching of new stay points is carried out, and analogized, the distance between the N th point Mn and the occupational space clustering center is calculated, meanwhile, for the time difference delta T between M1 and Mn-1, if the delta T is larger than the stay time threshold value T, and the normal travel time distribution is met, a single occupational travel is considered to be formed, the travel is identified as the occupational destination travel, a stay point set { N1, N2, … Nn } is generated for the points which do not meet the occupational core clustering, and feature analysis based on other travel purposes of the time space is carried out.
According to the invention, according to user signaling data and user internet surfing data which are triggered by high-frequency signals and have time-space correlation characteristics, a characteristic identification method for meeting different user travel activities with the time-space correlation characteristics is provided through the time-space correlation characteristics and time-space kernel cluster analysis of mobile phone positioning data, and based on the travel activity purpose identification method, OD matrixes of different user travel activities are finally generated, so that big data support meeting the time-space travel activity chain characteristics is provided for traffic planning and traffic demand management; the stagnation point judgment method of the invention conforms to the national law privacy regulations and has the following beneficial effects:
1) the method can meet the corresponding relationship of the job and the live and the travel characteristics of the user, and has the advantages of simple acquisition mode, low cost, large information sample, flexible sampling time, automatic acquisition and the like compared with the traditional traffic survey;
2) the method for identifying the characteristics of different travel purposes based on the time-space correlation characteristic analysis integrates and optimizes the identification method of the characteristics of the travel activities and the travel OD residence point generation method, and can effectively generate a one-day travel activity chain of a user, thereby being beneficial to the OD travel research of the travel activities and providing data support for the traffic planning management department to research different groups and different travel activities.
3) According to the method, the respective advantages of the time-space correlation OD matrix generation method and the time-space kernel clustering analysis are combined, the OD sequences with the time-space correlation characteristics and different trip activity target characteristics are finally formed, the reasonability and the accuracy of the OD matrix of the trip destination are further improved, and the cost is effectively reduced.
Drawings
FIG. 1 is a flow chart of a method for identifying characteristics of different trip activities based on mobile phone positioning data;
FIG. 2 user trajectories and stationing points for commuting travel purposes with job identification;
FIG. 3 user trajectories and waypoints for different travel purposes with occupational and livelihood destination designations;
FIG. 4 shows user trajectories and stops for other travel purposes with job identification.
Detailed Description
The following detailed description of the embodiments of the present invention is provided with reference to the accompanying drawings, and referring to fig. 1, the following steps are specifically performed:
step 1: and acquiring signaling data and internet surfing data of the mobile phone. Mobile phone signaling data and internet data samples such as table 1, wherein the mobile phone signaling data comprises user portrait label attribute data, and the user portrait label attribute data samples such as table 2;
table 1 sample of mobile phone signaling data and internet data
TABLE 2 user Profile tag Attribute data sample
Step 2: using the mobile phone signaling data and the internet surfing data collected in the step 1, extracting mobile phone triggering data within three months to judge the occupational position, determining the occupational position LiveLoc and the employment position WorkLoc, and obtaining occupational position distribution characteristics; the mobile phone trigger data comprises label attribute information such as a user number, a base station number, a timestamp, longitude, latitude, individual age, gender, whether students or not, whether retired persons or not and the like.
Step 2.1: judging the residence: first, mentionTaking mobile phone triggering data in three months in the step 1, sequencing the same mobile phone number time sequence, and defining the time range of residence judgment as 21:00 evening to 07:00 early morning; then carrying out spatial clustering on the mobile phone data with the same user number to form a spatial clustering point cluster Clus _ n; frequent item sets of different stay durations Tday based on spatial clustering point clusters Clus _ n
Analyzing, performing weight analysis by combining with the power-on and power-off event occurrence place of a user at night, and performing weight design according to the principle that the longer the retention time is, the more the occurrence times are, and the more the power-on and power-off event occurrence places are, the larger the weight is; then, according to the weighted value wi of the frequent item set, counting the weight residence probability P of the space clustering point cluster Clus _ nTsay_iThen, the maximum value P of the probability of the weight residence is calculatedTsay_i_maxWeighted probability maximum (P) in the spatial cluster of pointsTsay_i_max)Clus_nNamely the place of residence.
Step 2.2: and 6, employment judgment: firstly, based on mobile phone trigger data in three months sorted in the step 2.1, a time range of employment place judgment is defined to be between 9:00 and 16:00 in the daytime, retired old people and minors are eliminated according to user portrait label attribute data of an operator, and then employment place judgment is carried out on the rest of people; similar to the judgment of the residence, carrying out spatial clustering on the mobile phone data with the same user number to form a spatial clustering point cluster Clus _ d; frequent item sets of different stay durations Tday based on spatial clustering point clusters Clus _ d
Analyzing, and designing the weight according to the principle that the longer the retention time is, the more the occurrence times is, the larger the weight is; then, according to the weight value wi of the frequent item set, the probability P of the weight employment place of the spatial clustering point cluster Clus _ d is countedTsay_iCalculating the probability maximum value (P) of the weight employment place in the spatial clustering point clusterTsay_i_max)Clus_dIf the maximum weight employment probability (P)Tsay_i_max)Clus_d_centerGreater than a membership threshold, and, maximumWeighted place of employment probability (P)Tsay_i_max)Clus_d_centerIf the spatial clustering point cluster and the residence are not in the same cluster, the place is judged to be the employment place; otherwise, the employment location is not judged.
And step 3: determining OD (origin-destination) stay points of a spatio-temporal dimension based on a time sequence with stay characteristics, namely, finding stay points meeting spatio-temporal displacement conditions based on time, distance and speed of vehicle displacement by a heuristic search algorithm, matching the stay points with stay data of a user, performing spatio-temporal kernel clustering analysis based on spatio-temporal association on a point set with central clustering characteristics and spatio-temporal trip characteristics, marking the stay points meeting the stay distribution characteristics, and forming an OD set sequence with spatio-temporal association stay characteristics;
step 3.1: and (3) heuristic search algorithm: for track point data sorted by time of a certain user, for one of the stationing points S _ i, the distance D _ i _ i +1 from the i +1 th point is calculated, if the distance D _ i _ i +2 from the i +2 th point to the stationing point S _ i > D _ i _ i +1, the maximum distance maxDist is D _ i _ i +2, and so on, the farthest distance point is calculated to be D _ i _ i + n, if the distance from the stationing point S _ i to the n +1 th point is greater than the distance from the stationing point S _ i to the n +2 th point, and the distance from the stationing point S _ i to the n +1 th point is greater than the distance threshold, and the speed from the stationing point S _ i to the n +2 th point is less than the speed threshold, the stationing state is determined.
Step 3.2, the judgment method of the time-space kernel clustering resident points with the position and residence correlation characteristics comprises the following steps: based on the OD residence point set with spatio-temporal correlation characteristics, first, clustering analysis on spatial dimensions is performed, and if the distances from the user spatio-temporal residence points M1 or M2 to the occupational region clustering center are both smaller than the distance threshold d1, it is considered that M1 and M2 may constitute occupational region residence and travel. Recursion to the 3 rd point M3, space dimension matching of new stay points is carried out, and analogized, the distance between the N th point Mn and the occupational space clustering center is calculated, meanwhile, for the time difference delta T between M1 and Mn-1, if the delta T is larger than the stay time threshold value T, and the normal travel time distribution is met, a single occupational travel is considered to be formed, the travel is identified as the occupational destination travel, a stay point set { N1, N2, … Nn } is generated for the points which do not meet the occupational core clustering, and feature analysis based on other travel purposes of the time space is carried out.
And 4, step 4: based on the OD set sequence with the time-space associated job and live characteristics in the step 3, the live activity trip judgment of the time-space dimension is carried out, all residence points of the user base trip are extracted for live activity time-space trip analysis, the time-space frequent item set analysis of the live activity trip is carried out, the residence points meeting the live activity trip characteristics are marked, the live activity trip OD set is generated, the residence points not meeting the job and live activity trip characteristics are marked as other residence points, and the OD set sequence sharing other trip purposes is generated.
Step 4.1: performing space kernel clustering analysis and judgment on the bioactivity based on a time sequence with the occupation characteristics, performing space clustering on a time space OD resident point of a certain user when a user goes out of a base house, and reforming a space clustering point cluster Clus _ n; then, based on the spatial clustering point cluster Clus _ n, frequent item sets with different retention time Tday
Analyzing, and designing the weight according to the principle that the longer the retention time is, the more the occurrence times are, and the more frequent the interaction with the family is; then, according to the weight value wi of the frequent item set, the weight activity destination probability P of the space clustering point cluster Clus _ n is countedTsay_iThen, the maximum value P of the probability of the weight activity destination is calculatedTsay_i_maxWeighted probability maximum (P) in the spatial cluster of pointsTsay_i_max)Clus_nNamely the bioactive destination.
Step 4.2, judging the OD trip of the activity trip activity target with the space-time correlation characteristics: based on the OD residence point set with the time-space correlation characteristics and the identification of the liveness destination, firstly, on the space dimension, whether the distance of the residence point set meets the space distance threshold of the liveness destination or not is calculated from the residence point set { N1, N2, … Nn } in the step 3.2, then, on the time-space dimension, time sequence judgment is carried out, whether a certain residence time length is met or not, and the time distribution of the liveness trip is carried out, if the time-space correlation characteristics of the live trip are met, a live trip is considered to be formed, the trip is identified as the live trip, and other residence point data which do not meet the time-space correlation characteristics are identified as other trips.
And 5: and 4, carrying out functional area identification on the space-time travel OD stagnation points of different travel purposes formed in the step 4 by combining the current land utilization characteristics, further judging whether stagnation point set sequences meeting the time-space travel characteristics have correlation characteristics in the time space, and further carrying out time-space clustering to form a time-space stagnation point set with time-space activity characteristics.
Step 6: performing characteristic OD verification of different activity purposes: and (4) checking the characteristic trip chains of different trip activities obtained in the step (5), judging whether the algorithm meets the characteristic data of the crowd with different activity characteristics, if the algorithm meets the characteristics, the checking is qualified, otherwise, further optimizing the space-time trip characteristic analysis algorithm of different trip purposes in the step (3) and the step (4). And perform a special crowd activity feature analysis, such as user trajectories and stationing points for commuting travel purposes with job identification, as shown in fig. 2; user trajectories and stops for different travel purposes with occupational and livelihood destination designations, as shown in fig. 3; user trajectories and stops for other travel purposes with job identification are shown in fig. 4.
And 7: according to the traffic cell division scheme, an operator base station is matched with a traffic cell, the origin-destination point of each traveler is mapped to the corresponding traffic cell, and an OD matrix table of different travel activities is generated based on the stagnation point.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (6)
1. A feature recognition method for different trip activities based on mobile phone positioning data is characterized by comprising the following specific steps:
step 1: acquiring mobile phone signaling data and internet surfing data, wherein the mobile phone signaling data comprises user portrait label attribute data;
step 2: and (3) extracting mobile phone trigger data for three months by using the mobile phone signaling data and the internet surfing data collected in the step (1) to judge the position: firstly, carrying out spatial clustering on places with three months constant occurrence, carrying out high-frequency analysis according to the occurrence frequency of a stay time interval at the spatial clustering points, and carrying out occupational judgment on the high-frequency points meeting the time and space requirements, wherein the residence judgment is to be combined with the night constant on-off frequency to carry out frequency weight analysis judgment, the employment judgment is to be combined with a user age label to carry out frequent item set resident employment characteristic judgment based on duration, the residence and employment of a user are obtained, namely the occupation distribution characteristic is obtained, and then the residence distribution characteristic is matched with a time sequence to generate the time sequence with the occupation characteristic;
and step 3: based on the time sequence with the occupation characteristics in the step 2, determining OD (origin-destination) residence points of the spatio-temporal dimensions, namely, based on the time, distance and speed of vehicle displacement, finding out residence points meeting spatio-temporal displacement conditions through a heuristic search algorithm, matching the residence points with occupation data of a user, performing spatio-temporal kernel clustering analysis based on spatio-temporal association on a point set with central clustering characteristics and spatio-temporal trip characteristics, marking the residence points meeting occupation distribution characteristics, further performing spatio-temporal clustering analysis on the residence points not meeting occupation distribution characteristics, further determining whether the residence points belong to the residence points, and forming an OD set sequence with the spatio-temporal association occupation characteristics;
and 4, step 4: based on the OD set sequence with the time-space associated job and live characteristics in the step 3, judging the live activity trip of the time-space dimension, extracting all resident points of the user's home trip for live activity time-space trip analysis, performing time-space frequent item set analysis of the live activity trip, marking the resident points meeting the live activity trip characteristics to generate a live activity trip OD set, and marking the resident points not meeting the job distribution and the live activity trip characteristics as other resident points to generate an OD set sequence sharing other trip purposes;
in the step 4, the judgment of the liveness trip is specifically as follows: performing space kernel clustering analysis and judgment on the bioactivity based on a time sequence with the occupation characteristics, performing space clustering on a time space OD resident point of a certain user when a user goes out of a base house, and reforming a space clustering point cluster; analyzing frequent item sets with different residence time lengths based on the spatial clustering point clusters, and designing weights according to the principle that the longer the residence time is, the more the occurrence times are, and the more frequent the interaction with the family is; then, according to the weight values of the frequent item sets, counting the probability of the weight liveness destination of the spatial clustering point cluster, and then calculating the maximum value of the probability of the weight liveness destination, wherein the maximum value of the probability of the weight in the spatial clustering point cluster is the liveness destination;
in the step 4, the OD trip judgment of the liveness trip activity objective having the time-space correlation characteristic specifically includes: based on an OD (origin-destination) stay point set with a temporal-spatial correlation characteristic and an identification of a liveness destination, in a spatial dimension, whether the distance of a non-working stay point set meets a spatial distance threshold of the liveness destination or not is judged, meanwhile, in a temporal-spatial dimension, a time sequence is judged, whether a certain stay time length is met or not and liveness travel time distribution is met, if the temporal-spatial correlation characteristic of a life travel is met, a one-time liveness destination travel is formed, the travel is identified as a life travel for the current travel, and other stay point data which do not meet the temporal-spatial correlation characteristic are identified as other travels;
and 5: performing further time-space correlation analysis on the time-space OD set sequences of different trip activities obtained in the step (4), and further judging whether the time space has correlation characteristics or not by combining with the functional area identification of land utilization, and performing further time-space kernel clustering to form a time-space stationary point set with the time-space trip purposes;
step 6: dotting mobile phone signaling data of different trip characteristic crowds on a map to perform trip characteristic classification, completing verification of trip characteristics of the characteristic crowds, verifying the time-space stationing point set obtained in the step 5, judging whether the algorithm meets characteristics of different trip activities, if so, verifying to be qualified, otherwise, further optimizing the time-space trip characteristic analysis algorithm in the steps 3 and 4, and performing trip characteristic analysis on special crowds;
and 7: according to the traffic cell division scheme, the operator base station is matched with the traffic cell, the origin-destination point of each traveler is mapped to the corresponding traffic cell, and finally an OD matrix travel table based on different travel activities of the traffic cell is formed.
2. The method for identifying characteristics of different trip activities based on mobile phone positioning data according to claim 1, wherein: the mobile phone trigger data comprise a user number, a timestamp, longitude, latitude, individual age, gender, whether students or not and whether retired persons or not.
3. The method for identifying characteristics of different travel activities based on mobile phone positioning data according to any one of claims 1-2, characterized in that: in the step 2, the residence judgment specifically comprises: firstly, carrying out time sequence sequencing on the same mobile phone number on extracted mobile phone trigger data within three months, and defining the time range of residence judgment as 21:00 in the evening to 07:00 in the morning; then carrying out spatial clustering on the mobile phone data with the same user number to form a spatial clustering point cluster; analyzing frequent item sets with different stay durations based on the spatial clustering point clusters, performing weight analysis by combining with the power-on and power-off event occurrence places of users at night, and performing weight design according to the principle that the longer the stay time is, the more the occurrence times are, and the more the power-on and power-off event occurrence places are, the larger the weight is; and then, according to the weight values of the frequent item sets, counting the weight residence probability of the spatial clustering point clusters, and calculating the maximum value of the weight residence probability in the spatial clustering point clusters, wherein the maximum value of the weight probability in the spatial clustering point clusters is the residence.
4. The method for identifying characteristics of different travel activities based on mobile phone positioning data according to any one of claims 1-3, characterized in that: in the step 2, the employment judgment specifically comprises: firstly, carrying out time series sequencing on the same mobile phone number on extracted mobile phone trigger data in half a year, defining the time range of employment place judgment as 9:00 in the daytime to 16:00 in the afternoon, eliminating retired old people and minors according to the attribute data of the user portrait label of an operator, and then judging the employment places of the rest of people; carrying out spatial clustering on the mobile phone data with the same user number to form a spatial clustering point cluster; analyzing frequent item sets with different stay durations based on the spatial clustering point clusters, and carrying out weight design according to the principle that the longer the stay time is, the more the occurrence times are, the larger the weight is, and if the maximum probability of the weighted employment place is greater than a membership threshold value and the corresponding spatial clustering point cluster and the residence place are not in the same cluster, judging the occupation place; otherwise, the position is not judged as the employment place position.
5. The method for identifying characteristics of different travel activities based on mobile phone positioning data according to any one of claims 1-4, characterized in that: the heuristic search algorithm in the step 3 specifically comprises the following steps: for track point data sorted by time of a certain user, for one of the stationing points S _ i, the distance D _ i _ i +1 from the i +1 th point is calculated, if the distance D _ i _ i +2 from the i +2 th point to the stationing point S _ i > D _ i _ i +1, the maximum distance maxDist is D _ i _ i +2, and so on, the farthest distance point is calculated to be D _ i _ i + n, if the distance from the stationing point S _ i to the n +1 th point is greater than the distance from the stationing point S _ i to the n +2 th point, and the distance from the stationing point S _ i to the n +1 th point is greater than the distance threshold, and the speed from the stationing point S _ i to the n +2 th point is less than the speed threshold, the stationing state is determined.
6. The method for identifying characteristics of different travel activities based on mobile phone positioning data according to any one of claims 1-5, characterized in that: the method for judging the time-space kernel clustering resident points with the position and residence correlation characteristics in the step 3 comprises the following steps: based on an OD (origin-destination) stay point set with a spatio-temporal correlation characteristic, firstly, performing clustering analysis on a spatio-temporal dimension, and if distances from a user spatio-temporal stay point M1 or a user spatio-temporal stay point M2 to a place-of-employment clustering center are smaller than a distance threshold d1, considering that the user spatio-temporal stay point M1 and the user spatio-temporal stay point M2 are likely to form a place-of-employment stay and a trip; recursion to the 3 rd point M3, space dimension matching of new stay points is carried out, and analogized, the distance between the N th point Mn and the occupational space clustering center is calculated, meanwhile, for the time difference delta T between M1 and Mn-1, if the delta T is larger than the stay time threshold value T, and the normal travel time distribution is met, a single occupational travel is considered to be formed, the travel is identified as the occupational destination travel, a stay point set { N1, N2, … Nn } is generated for the points which do not meet the occupational core clustering, and feature analysis based on other travel purposes of the time space is carried out.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011211769.XA CN112215666A (en) | 2020-11-03 | 2020-11-03 | Characteristic identification method for different trip activities based on mobile phone positioning data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011211769.XA CN112215666A (en) | 2020-11-03 | 2020-11-03 | Characteristic identification method for different trip activities based on mobile phone positioning data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112215666A true CN112215666A (en) | 2021-01-12 |
Family
ID=74058109
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011211769.XA Pending CN112215666A (en) | 2020-11-03 | 2020-11-03 | Characteristic identification method for different trip activities based on mobile phone positioning data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112215666A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112579915A (en) * | 2021-02-26 | 2021-03-30 | 深圳市城市交通规划设计研究中心股份有限公司 | Analysis method and device for trip chain |
| CN112685659A (en) * | 2021-03-19 | 2021-04-20 | 上海钐昆网络科技有限公司 | Target location determination method and device, electronic equipment and computer storage medium |
| CN112836121A (en) * | 2021-01-28 | 2021-05-25 | 北京市城市规划设计研究院 | Travel purpose identification method and system |
| CN113377845A (en) * | 2021-08-12 | 2021-09-10 | 成都天府市民云服务有限公司 | Intelligent old-age care data processing method and system based on big data |
| CN113935881A (en) * | 2021-12-16 | 2022-01-14 | 北京融信数联科技有限公司 | Population structure analysis method and system based on big data and readable storage medium |
| CN115545759A (en) * | 2022-09-27 | 2022-12-30 | 暨南大学 | Passenger travel intention mining method based on dynamic pricing mechanism |
| CN116233823A (en) * | 2023-05-10 | 2023-06-06 | 深圳市城市交通规划设计研究中心股份有限公司 | Identification method of cross-city commute ring, electronic equipment and storage medium |
| CN116957520A (en) * | 2023-09-20 | 2023-10-27 | 北京融信数联科技有限公司 | Big data-based loss of business rate monitoring method, system and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105142106A (en) * | 2015-07-29 | 2015-12-09 | 西南交通大学 | Traveler home-work location identification and trip chain depicting method based on mobile phone signaling data |
| CN107194525A (en) * | 2017-03-23 | 2017-09-22 | 同济大学 | A kind of down town appraisal procedure based on mobile phone signaling |
| CN108492565A (en) * | 2018-04-20 | 2018-09-04 | 广东亿迅科技有限公司 | Public transport control method and system based on the analysis of user's trip data |
| CN108650632A (en) * | 2018-04-28 | 2018-10-12 | 广州市交通规划研究院 | It is a kind of based on duty live correspondence and when space kernel clustering stationary point judgment method |
| CN109005515A (en) * | 2018-09-05 | 2018-12-14 | 武汉大学 | A method of the user behavior pattern portrait based on motion track information |
| CN110135679A (en) * | 2019-04-01 | 2019-08-16 | 同济大学 | A kind of appraisal procedure based on mobile phone signaling data Living Center space performance |
| CN111737605A (en) * | 2020-07-09 | 2020-10-02 | 南京瑞栖智能交通技术产业研究院有限公司 | Travel purpose identification method and device based on mobile phone signaling data |
-
2020
- 2020-11-03 CN CN202011211769.XA patent/CN112215666A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105142106A (en) * | 2015-07-29 | 2015-12-09 | 西南交通大学 | Traveler home-work location identification and trip chain depicting method based on mobile phone signaling data |
| CN107194525A (en) * | 2017-03-23 | 2017-09-22 | 同济大学 | A kind of down town appraisal procedure based on mobile phone signaling |
| CN108492565A (en) * | 2018-04-20 | 2018-09-04 | 广东亿迅科技有限公司 | Public transport control method and system based on the analysis of user's trip data |
| CN108650632A (en) * | 2018-04-28 | 2018-10-12 | 广州市交通规划研究院 | It is a kind of based on duty live correspondence and when space kernel clustering stationary point judgment method |
| CN109005515A (en) * | 2018-09-05 | 2018-12-14 | 武汉大学 | A method of the user behavior pattern portrait based on motion track information |
| CN110135679A (en) * | 2019-04-01 | 2019-08-16 | 同济大学 | A kind of appraisal procedure based on mobile phone signaling data Living Center space performance |
| CN111737605A (en) * | 2020-07-09 | 2020-10-02 | 南京瑞栖智能交通技术产业研究院有限公司 | Travel purpose identification method and device based on mobile phone signaling data |
Non-Patent Citations (2)
| Title |
|---|
| 姚振兴: "基于手机传感器数据的个体交通出行链信息采集方法研究", 《中国博士学位论文全文数据库 (工程科技Ⅱ辑)》 * |
| 李耀辉: "基于移动信令数据的用户出行行为研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112836121A (en) * | 2021-01-28 | 2021-05-25 | 北京市城市规划设计研究院 | Travel purpose identification method and system |
| CN112579915A (en) * | 2021-02-26 | 2021-03-30 | 深圳市城市交通规划设计研究中心股份有限公司 | Analysis method and device for trip chain |
| CN112685659A (en) * | 2021-03-19 | 2021-04-20 | 上海钐昆网络科技有限公司 | Target location determination method and device, electronic equipment and computer storage medium |
| CN113377845A (en) * | 2021-08-12 | 2021-09-10 | 成都天府市民云服务有限公司 | Intelligent old-age care data processing method and system based on big data |
| CN113377845B (en) * | 2021-08-12 | 2021-10-29 | 成都天府市民云服务有限公司 | Intelligent old-age care data processing method and system based on big data |
| CN113935881A (en) * | 2021-12-16 | 2022-01-14 | 北京融信数联科技有限公司 | Population structure analysis method and system based on big data and readable storage medium |
| CN115545759A (en) * | 2022-09-27 | 2022-12-30 | 暨南大学 | Passenger travel intention mining method based on dynamic pricing mechanism |
| CN115545759B (en) * | 2022-09-27 | 2023-05-19 | 暨南大学 | Passenger trip intention mining method based on dynamic pricing mechanism |
| CN116233823A (en) * | 2023-05-10 | 2023-06-06 | 深圳市城市交通规划设计研究中心股份有限公司 | Identification method of cross-city commute ring, electronic equipment and storage medium |
| CN116233823B (en) * | 2023-05-10 | 2023-08-01 | 深圳市城市交通规划设计研究中心股份有限公司 | Identification method of cross-city commute ring, electronic equipment and storage medium |
| CN116957520A (en) * | 2023-09-20 | 2023-10-27 | 北京融信数联科技有限公司 | Big data-based loss of business rate monitoring method, system and storage medium |
| CN116957520B (en) * | 2023-09-20 | 2023-12-26 | 北京融信数联科技有限公司 | Big data-based loss of business rate monitoring method, system and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112215666A (en) | Characteristic identification method for different trip activities based on mobile phone positioning data | |
| CN108650632B (en) | Stagnation point judgment method based on occupational correspondence and time-space kernel clustering | |
| CN106600960B (en) | Travel origin-destination point identification method based on space-time clustering analysis algorithm | |
| CN106096631B (en) | A kind of floating population's Classification and Identification analysis method based on mobile phone big data | |
| CN110245981B (en) | Crowd type identification method based on mobile phone signaling data | |
| CN109495856B (en) | Mobile phone user type marking method based on big data | |
| CN108320501B (en) | Bus route identification method based on user mobile phone signaling | |
| Rong et al. | Du-parking: Spatio-temporal big data tells you realtime parking availability | |
| CN111582948B (en) | Individual behavior analysis method based on mobile phone signaling data and POI (Point of interest) | |
| Pei et al. | A new insight into land use classification based on aggregated mobile phone data | |
| CN106931974B (en) | Method for calculating personal commuting distance based on mobile terminal GPS positioning data record | |
| CN109089314B (en) | Indoor positioning method of wifi sequence assisted GPS based on recommendation algorithm | |
| CN111737605A (en) | Travel purpose identification method and device based on mobile phone signaling data | |
| Lee et al. | Urban spatiotemporal analysis using mobile phone data: Case study of medium-and large-sized Korean cities | |
| CN111212383B (en) | Method, device, server and medium for determining number of regional permanent population | |
| CN108062857B (en) | Prediction technique for cab-getter's trip purpose | |
| CN109684373B (en) | Key relation person discovery method based on travel and call ticket data analysis | |
| CN110414732A (en) | A kind of trip Future Trajectory Prediction method, apparatus, storage medium and electronic equipment | |
| Zhu et al. | Learning transportation annotated mobility profiles from GPS data for context-aware mobile services | |
| CN111144452A (en) | Mobile user trip chain extraction method based on signaling data and clustering algorithm | |
| CN114969007A (en) | Urban functional area identification method based on function mixing degree and integrated learning | |
| CN115168529A (en) | Hub passenger flow tracing method based on mobile phone positioning data | |
| CN117056823A (en) | Method and system for identifying occupation type of shared bicycle commuter user | |
| Leca et al. | Significant location detection & prediction in cellular networks using artificial neural networks | |
| CN112949784B (en) | Resident trip chain model construction method and resident trip chain acquisition method |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210112 |
|
| RJ01 | Rejection of invention patent application after publication |