CN107770744A - The identification of travelling OD node and hop extracting method under big data environment - Google Patents
The identification of travelling OD node and hop extracting method under big data environment Download PDFInfo
- Publication number
- CN107770744A CN107770744A CN201710843841.2A CN201710843841A CN107770744A CN 107770744 A CN107770744 A CN 107770744A CN 201710843841 A CN201710843841 A CN 201710843841A CN 107770744 A CN107770744 A CN 107770744A
- Authority
- CN
- China
- Prior art keywords
- node
- user
- time
- point
- space
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/20—Services signaling; Auxiliary data signalling, i.e. transmitting data via a non-traffic channel
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2474—Sequence data queries, e.g. querying versioned data
Abstract
The invention provides the identification of travelling OD node and hop extracting method under a kind of big data environment, using mobile terminal individual at the appointed time in the range of space operation data set, excavate the trip track data of a large amount of individuals, interpolation is fitted to it, obtains the individual trip Time-space serial of constant duration；Possible cluster areas is searched in individual goes on a journey Time-space serial using spatial clustering method, and whether the cluster point for judging to extract by comparing the angle of cut difference of cluster areas central point and cluster areas external node is OD points, and Time-space serial of being gone on a journey to user is split.The present invention can low cost, automation, the trip Time-space serial for easily obtaining a large amount of populations in the range of specified time, utilization space clustering algorithm and weighted average method, rapidly find out the node region with OD features, OD point differentiations are carried out according to rule, the section segmentation based on OD nodes is carried out so as to trip Time-space serial convenient, efficiently to user.
Description
Technical field
The present invention relates to the travelling OD under a kind of big data environment, based on magnanimity anonymity encryption times sequence location data
Node identifies and OD hop extracting methods, belongs to big data analysis technical field.
Background technology
In recent years, as explosive growth is presented in the development of information technology, data message amount, data source is more and more,
Data volume is also more and more huger.Wherein, the data recorded by information sensors such as mobile phone, WIFI, Internet of Things have become big number
According to most important data source in analysis, its more complete individual trip is recorded as big data, especially traffic big data point
Analysis provides good data and supported.By taking mobile phone as an example, to 2015, cellphone subscriber reached 13.06 hundred million, accounts for total population
More than 96%, mobile phone terminal equipment continues caused signal message, forms the volume of data collection of record user's trip, is
The behaviors such as the trip of analysis Urban population, delay provide important data source.
However, the mobile phone signaling big data for being available for obtaining at present contains only anonymous encryption user and the communication of base station is remembered
Record, wherein related to user's travel behaviour only call duration time and base station number etc., the travel behaviour of user (including go on a journey
Terminal, middle stop ground, trip route, trip mode etc.) simply lie in mobile phone signaling, not directly performance
Out, this just needs a kind of efficient, succinct algorithm to be gone on a journey to the user that is formed with mobile phone signaling data at track data
Reason, the O-D points of user's trip are identified, split the O-D paths of user's trip, extract the travel behaviour feature of user.Existing skill
In art, not such algorithm.
The content of the invention
It is used for rail of being gone on a journey to the user formed with mobile phone signaling data the technical problem to be solved in the present invention is to provide a kind of
Mark data are handled, and identify the O-D points of user's trip, the method in the O-D paths of segmentation user's trip.
In order to solve the above-mentioned technical problem, the technical scheme is that providing travelling OD section under a kind of big data environment
Point identification and hop extracting method, it is characterised in that this method is made up of following 5 steps：
Step 1, the anonymous encryption mobile terminal sensing data obtained from sensor operator is read, anonymity encryption moves
Dynamic terminal sensor data is continuous in the time and space in theory, and different mobile terminal corresponds to different user's unique numbers
EPID, the communication signaling record that each EPID is at the appointed time triggered in section is extracted, form the trip data collection of the EPID；
Step 2, communications records of each EPID at the appointed time in section with sensor are extracted successively, arrange in chronological order
Sequence, establish user's trip track data collection；From start time t0, using locus of the T time as interval to trip data
Enter row interpolation, build the user's trip Time-space serial being made up of true point and interpolation point；
Most trifle in step 3, Time-space serial of being gone on a journey to user in the node clustering density p and O-D point radiuses of O-D points
Point quantity ε, and the angle of cut of the front and rear trip Time-space serial of O-D points are analyzed, the initialization of arrange parameter, while formulate O-D
The decision rule of point；
Step 4, based on DBSCAN algorithms, design a kind of Spatial Clustering based on distance；Obtained in step 2
On the basis of clustering parameter, space clustering is carried out to the interpolation point that user goes on a journey in Time-space serial data, extracts the poly- of egress
Class cluster, therefrom extract the most crucial point in clustering cluster；Calculate from most crucial point, first to before and after node clustering cluster
The deflection of individual node, calculate the angle of cut of trip Time-space serial of the user before and after node cluster；
Step 5, the decision rule obtained according to step 2, when whether the clustering cluster that discriminating step 4 obtains is user's trip
O-D points on empty sequence, the O-D clustering clusters obtained to differentiation arrange, and using its most crucial node as O-D points, search it
The measuring point of original close positions, its position is projected on map, record its actual position；According to extracting
The trip data of the O-D points segmentation user arrived, user's trip track being segmented, having shown starting point.
Preferably, the step 1 includes：
Step 1.1, read from the anonymous encryption mobile terminal sensing data of sensor operator acquisition, it is anonymous in theory
It all should be continuous in the time and space to encrypt mobile terminal sensing data, including：User's unique number EPID, lead to
Believe type of action TYPE, communication operation moment TIME occurs, great Qu REGIONCODE, sensing implement body numbering residing for sensor
SENSORID；Wherein, great Qu REGIONCODE residing for sensor and sensing implement body numbering SENSORID constitute sensor volume
Number；
Step 1.2, an anonymous encryption mobile terminal sensing data are that a signaling records, and every signaling is recorded
It is decrypted；
Step 1.3, according to user unique number EPID, inquire about its at the appointed time log all in section, structure
User's trip track data.
It is highly preferred that the step 2 includes：
All fixation sensor numbers in user's trip track data that step 2.1, extraction step 1.3 obtain
Latitude and longitude coordinates are converted to geographical coordinate X- by REGIONCODE-SENSORID and its corresponding latitude and longitude coordinates LON-LAT
Y；
Step 2.2, traverse user trip track data, it is arranged by triggering call duration time TIMESTAMP orders；
Step 2.3, trip data is begun stepping through from start time, adjacent every 3 communications records point fitting one is secondary
Curve, the x-axis of conic section is the time of user's trip track, and y-axis is the X-Y coordinates of communications records point；If the trip of user
Track includes n communications records point, and n is positive integer, then needs to fit 2n-4 bar conic sections altogether；
Step 2.4, from integer start time t0, T calculates X-Y of the user at each time point and sat at timed intervals
((t0+nT) forms an interpolation point, in addition to two sections of head and the tail, is all deposited between remaining communication point by mark, same time X (t0+nT) and Y
In 2 matched curves, the X-Y coordinate of interpolation point between is averaging to obtain by the result of calculation of two curves；
Step 2.5, all interpolation points sort in chronological order, and all interpolation points form user's trip Time-space serial.
Preferably, the step 3 includes：
Step 3.1, the user's trip Time-space serial for choosing O-D points in some clear and definite traces form sample as sample
Time-space serial, analyze and mark its O-D region, O-D regions are O-D node sets；
Step 3.2, traversal sample Time-space serial, find out first node and last section in each O-D regions
Point, it is assumed that have n1 node, n1 is positive integer, and since first node, segmentation calculates the space between O-D Area Nodes
These distances are summed up by distance, a total of n1-1 sections, are calculated O-D Area Node density ps and are
Step 3.3, the weighted center for calculating O-D regions, weight w are O-D regional nodes apart from its central point that sorts
Sequence number is poor, if O-D regional nodes number is n1, the weight w minimums of the 1st and the n-th 1 node, theIndividual node
Weight it is maximum, the central point C in whole O-D regions X-Y coordinates (Xc、Yc) be：
Step 3.4, the coordinate according to first point F1, B1 before and after the central point C and O-D regions in O-D regions, by F1-
C is connected with C-B1, it is assumed that F1 coordinates are (XF, YF), and B1 coordinates are (XB, YB), and C coordinates are (XC, YC), it is known that three point coordinates
Seek F1-C and C-B1 angle of cut difference ∠ C：
Wherein, DFCFor the distance between node F1 and node C, DCBFor the distance between node C and node B1, DFBFor section
The distance between point F1 and node B1；
Step 3.5, the analysis according to Time-space serial of being gone on a journey to sample, what obtained node space cluster and O-D points differentiated
The nodal distance density and node number of parameter, i.e. O-D regions, and user's travel path is in the angle of turning back in O-D points region
Poor size, average value processing is carried out to the analysis result of sample data, obtains the criterion of batch O-D points identification, i.e. O-D areas
The threshold limit value Thr- ρ and node number threshold limit value Thr- ε of the nodal distance density in domain, and in the feelings for being unsatisfactory for Thr- ρ
Under condition, O-D points both ends stroke is turned back differential seat angle Thr- ∠.
It is highly preferred that in the step 3.2, the calculation formula of O-D Area Node density psIn, molecule subtracts
1 be ensure section number it is consistent with nodes, denominator add 1 be prevent 0 situation.
It is highly preferred that the step 4 includes：
Step 4.1, the currently pending user's trip Time-space serial handled by step 2 is read from database
Data, begun stepping through from start time, centered on each node, search the Thr- ε neighbor nodes in its front and rear neighborhood,
Front and rear neighborhood is eachIt is individual；
Step 4.2, assume that it is positive integer currently to traverse node n2, n2, then the border of its front and rear point of proximity is nodeWithCount from nodeTo nodeSegmentation distance plus andWherein,According to step 3.2, calculate under node number threshold limit value Thr- ε, using node n2 as
The node density of the neighborhood at center；
Step 4.3, judge whether the node density of the neighborhood centered on point n2 is more than threshold limit value Thr- ρ；
It is reachable for density in the node field if being more than Thr- ρ, the neighborhood is labeled as a quasi- node clustering cluster,
Make its density be connected, its internal all node is all marked into the cluster node that is defined, records a node before and after its outside
Space coordinates；
If being less than Thr- ρ, give up the neighbor domain of node, continue to travel through next node；
Step 4.4, when traveled through user go on a journey Time-space serial after, travel through the quasi- node clustering each judged from the beginning
Cluster, judge whether there is common factor between the node that adjacent quasi- node clustering is included；Occur simultaneously if existing, claim this two quasi- sections
Density is reachable between point clustering cluster, and the two quasi- node clustering clusters are merged into one, make its density be connected, recalculate this
The node density of quasi- node clustering cluster after merging；
If new node density is more than Thr- ρ, confirms to merge the two quasi- node clustering clusters, record new conjunction again
And quasi- node clustering cluster outside before and after first node space coordinates；
If new node density is less than Thr- ρ, the quasi- node clustering cluster of merging is taken apart again, and meet that node is close
Spend under conditions of Thr- ρ and node number Thr- ε, abandoned one by one in the presence of the node to occur simultaneously in former quasi- node clustering cluster；
The each quasi- node clustering cluster of step 4.5, traversal, the central point of each quasi- node clustering cluster is calculated according to step 3.3
C, after central point C is obtained, the X-Y of first point is sat before and after each quasi- node clustering cluster outside obtained according to step 4.4
Mark, calculate user gone on a journey before and after C points track the angle of cut it is poor, and recorded.
Further, the step 5 includes：
Quasi- node clustering cluster in every step 5.1, traversal user's trip Time-space serial, calculates what is obtained according to step 4
The angle of cut of node density and front and rear node is poor, according to step 3 obtain decision rule, judge its whether be user trip when
O-D points on empty sequence, if it is determined that being O-D points, then the cluster is labeled as node clustering cluster；
Step 5.2, the node clustering cluster identified is marked in user goes on a journey Time-space serial, give up the institute in clustering cluster
There is node, the central point C obtained with step 4.5 is replaced, as O-D points；
Step 5.3, by all O-D spot projections to map, obtain O-D points on map using space correlation method
Real space location name；
Step 5.4, using O-D as end points split user go on a journey Time-space serial, obtain divide starting point user go on a journey track.
The invention provides a kind of passenger flow O-D identifications based on magnanimity anonymity encryption times sequence location data and O-D roads
The method of footpath extraction, the individual trip track of magnanimity is built according to the time of individual and spatial position data；Pass through interpolation method
Node on trip track is expanded；The processing parameter of user's trip moment sequence is obtained by sample data；By right
Spatially position is clustered node in individual trip Time-space serial data, and pavement branch sections are segmented out for end points with cluster；According to
The nodes of angle and cluster between single O-D sections judge whether the cluster is O-D points；Pass through the O-D identified
Point, the O-D sections of extraction user's trip.
The communication that method provided by the invention is leveraged fully between the mobile terminal and sensor that existing user holds is big
Data resource, using the lasting encryption position information of existing magnanimity anonymity mobile terminal in communication network, can low cost, from
Dynamicization, the trip Time-space serial for easily obtaining a large amount of populations in the range of specified time, therefrom select sample sequence and carry out parameter
With the training of rule, utilization space clustering algorithm and weighted average method, rapidly find out user's trip space-time based on this
In sequence carry O-D features node region, according to rule carry out O-D point differentiations, so as to it is convenient, efficiently user is gone out
Row Time-space serial carries out the section segmentation based on O-D nodes.
Brief description of the drawings
Fig. 1 is the identification of travelling OD node and hop extracting method stream under the big data environment that the present embodiment provides
Cheng Tu.
Embodiment
With reference to specific embodiment, the present invention is expanded on further.
Fig. 1 is the identification of travelling OD node and hop extracting method stream under the big data environment that the present embodiment provides
Cheng Tu, the identification of travelling OD node and hop extracting method are moved using anonymous encryption under described big data environment
At the appointed time (i.e. mobile terminal individual is with fixing sensor for the activity data collection in scope and spatial dimension for terminal individual
Communications records), user's trip track is formed, row interpolation expanding node is entered to trip track, forms user's trip Time-space serial,
On the basis of key parameter is obtained by sample data, it is intensive that Time-space serial interior joint is extracted using the method for space clustering
Region, mark off pavement branch sections, by comparing the angle between pavement branch sections, judgement identifies that user goes on a journey in moment sequence
O-D points, O-D paths are split with this.
This method comprises the following steps that：
Step 1, system read from sensor operator and obtain anonymous encryption mobile terminal sensing data, anonymity encryption
Mobile terminal sensing data is continuous in the time and space in theory, and different mobile terminal corresponds to different EPID, and extraction is every
The communication signaling record that individual EPID is at the appointed time triggered in section, form the trip data collection of the EPID；
Anonymity encryption mobile terminal sensing data is operator from mobile communications network, fixed broadband network, wireless
WIFI and location-based service correlation APP etc. are obtained in real time and the encrypted bits for the anonymous cellphone subscriber's time series after encrypting that desensitizes
Confidence ceases, and content includes：EPID, TYPE, TIME, REGIONCODE, SENSORID, referring to Application No.
201610273693.0 Chinese patent.It is specifically described as follows：
EPID (anonymous One-Way Encryption whole world unique mobile terminal identification code, EncryPtion international
Mobile subscriber IDentity), it is that unidirectional irreversible encryption is carried out to each mobile terminal user, so as to unique
Each mobile terminal user is identified, and does not expose Subscriber Number privacy information, it is desirable to after each mobile terminal user's encryption
EPID keeps uniqueness, i.e. the EPID of any time each cellphone subscriber keeps constant and do not repeated with other cellphone subscribers.
TYPE, it is the communication operation type involved by current record, e.g., online, call, calling and called, transmitting-receiving short message, GPS
Positioning, the switching of sensor cell, sensor switching, switching on and shutting down etc..
TIME, it is that the moment occurs for the communication operation involved by current record, unit is millisecond.
REGIONCODE, SENSORID are the sensor encrypted bits confidences that the communication operation involved by current record occurs
Breath.The numbering of REGIONCODE, SENSORID sensor, great Qu, SENSORID wherein residing for REGIONCODE representative sensors
It is the numbering of specific sensor.
Step 1.1, system read from sensor operator and obtain anonymous encryption mobile terminal sensing data, in theory
Anonymity encryption mobile terminal sensing data all should be continuous in the time and space, including：User's unique number
Moment TIME, great Qu REGIONCODE, sensing utensil residing for sensor occur for EPID, communication operation type TYPE, communication operation
Body numbering SENSORID；Wherein, great Qu REGIONCODE residing for sensor and sensing implement body numbering SENSORID constitute biography
Sensor is numbered；
Step 1.2, an anonymous encryption mobile terminal sensing data are that a signaling records, and every signaling is recorded
It is decrypted；
Step 1.3, according to Customs Assigned Number EPID, inquire about its at the appointed time log all in section, build user
Trip data；
In the present embodiment, the real-time signaling record data for extracting obtained user and sensor is：
The real-time signaling record data that table 1 newly receives after decrypting
| RECORDID | EPID | TYPE | TIMESTAMP | REGIONCODE | SENSORID |
| …… | …… | …… | …… | …… | …… |
| R501 | E1 | T1 | 2017-05-10 10：05：43 | 9878 | 3415 |
| R502 | E1 | T2 | 2017-05-10 10：21：14 | 9878 | 3418 |
| R503 | E1 | T1 | 2017-05-10 10：39：26 | 9878 | 4632 |
| R504 | E1 | T2 | 2017-05-10 10：52：19 | 9878 | 6343 |
| R505 | E1 | T1 | 2017-05-10 11：07：23 | 9878 | 1242 |
| R506 | E1 | T2 | 2017-05-10 11：15：34 | 9878 | 1253 |
| R507 | E1 | T2 | 2017-05-10 11：35：41 | 9878 | 1253 |
| R508 | E1 | T2 | 2017-05-10 11：47：14 | 9878 | 1253 |
| R509 | E1 | T2 | 2017-05-10 12：05：32 | 9880 | 3423 |
| R510 | E1 | T2 | 2017-05-10 12：17：13 | 9880 | 1454 |
| R511 | E1 | T1 | 2017-05-10 12：31：42 | 9880 | 9876 |
| R512 | E1 | T1 | 2017-05-10 12：44：21 | 9880 | 7645 |
| R513 | E1 | T1 | 2017-05-10 12：55：32 | 9880 | 7645 |
| R514 | E1 | T1 | 2017-05-10 13：06：12 | 9880 | 2342 |
| R515 | E1 | T1 | 2017-05-10 13：15：55 | 9880 | 5423 |
| R516 | E1 | T2 | 2017-05-10 13：31：24 | 9880 | 4643 |
| …… | …… | …… | …… | …… | …… |
The positional information of step 2, all fixed sensors of extraction, carries out Cluster merging processing to it, then calculates it
Voronoi diagram, the actual control range (i.e. its Thiessen polygon) of each sensor is obtained, records each Thiessen
The area of polygon, set the three-level population bearing capacity of each Thiessen polygons；
The user that step 2.1, extraction step 1.3 obtain goes on a journey in track data, all fixation sensor numbers
Latitude and longitude coordinates are converted to geographical coordinate X- by REGIONCODE-SENSORID and its corresponding latitude and longitude coordinates LON-LAT
Y；
In the present embodiment, the numbering of fixed sensor and geographical coordinate are shown in Table 2：
Fixation sensors X-Y-coordinate after the conversion of the longitude and latitude of table 2
Step 2.2, traverse user trip track data, it is arranged by triggering call duration time TIMESTAMP orders；
Step 2.3, trip data is begun stepping through from start time, adjacent every 3 communications records point fitting one is secondary
Curve, the x-axis of conic section are gone on a journey time of track for user, y-axis is the X-Y coordinates of communications records point, if so user
Trip track includes n communications records point, and n is positive integer, then needs to fit 2n-4 bar conic sections altogether；
Step 2.4, from integer start time t0, T calculates X-Y of the user at each time point and sat at timed intervals
((t0+nT) forms an interpolation point, in addition to two sections of head and the tail, is all deposited between remaining communication point by mark, same time X (t0+nT) and Y
In 2 matched curves, the X-Y coordinate of interpolation point between is averaging to obtain by the result of calculation of two curves；
In the present embodiment, it is 10 to make start time t0：05, time interval T are 10 minutes, the user obtained after interpolation
Trip Time-space serial is shown in Table 3.
The interpolated data of table 3 and record data
Step 2.5, all interpolation points sort in chronological order, and all interpolation points form the trip Time-space serial number of user
According to；
In the present embodiment, after removing communication node, the user being only made up of interpolation point goes on a journey Time-space serial to be shown in Table 4.
The interpolated data of table 4 and record data
Most trifle in step 3, space-time track of being gone on a journey to sample in the node clustering density p and O-D point radiuses of O-D points
Point quantity ε, and the angle of cut (differential seat angle) of the front and rear trip Time-space serial of O-D points are analyzed, the initialization of arrange parameter, together
When formulate O-D points decision rule；
Step 3.1, the user's trip Time-space serial for choosing O-D points in some clear and definite traces are analyzed and marked as sample
Go out its O-D region (O-D node sets)；
In the present embodiment, the sample data of use and the identification of O-D points are shown in Table 5a and table 5b.
Table 5a user trip Time-space serial sample
Table 5b user trip Time-space serial sample
| RECORDID | TIMESTAMP | X | Y | O-D |
| …… | …… | …… | …… | …… |
| INS32 | 2017-05-05 14：35：00 | 6256.6045 | 6460.7351 | F |
| INS33 | 2017-05-05 14：40：00 | 6457.6882 | 6402.5267 | F |
| INS34 | 2017-05-05 14：45：00 | 6669.3553 | 6339.0265 | F |
| INS35 | 2017-05-05 14：50：00 | 6875.7307 | 6270.2347 | T |
| INS36 | 2017-05-05 14：55：00 | 6875.7307 | 6270.2347 | T |
| INS37 | 2017-05-05 15：00：00 | 6875.7307 | 6270.2347 | T |
| INS38 | 2017-05-05 15：05：00 | 6738.1471 | 6434.2767 | T |
| INS39 | 2017-05-05 15：10：00 | 6526.4800 | 6434.2767 | F |
| INS40 | 2017-05-05 15：15：00 | 6314.8129 | 6418.4017 | F |
| …… | …… | …… | …… | …… |
By sample Time-space serial, first node for finding out each O-D regions saves with last for step 3.2, traversal
Point, it is assumed that have n1 node, n1 is positive integer, and since first node, segmentation calculates the space between O-D Area Nodes
These distances are summed up by distance, a total of n1-1 sections, are calculated O-D Area Nodes density (Distance Density) ρ and are
Wherein, molecule, which subtracts 1 and is, ensures that section number is consistent with nodes, denominator add 1 be prevent 0 situation；
In the present embodiment, the node density in the O-D regions in table 5a and table 5b is respectively 0.0195 and 0.0139.
Step 3.3, the weighted center for calculating O-D regions, weight w are O-D regional nodes apart from its central point that sorts
Sequence number is poor, if O-D regional nodes number is n1, the weight w minimums of the 1st and the n-th 1 node, theIndividual node
Weight it is maximum, the central point C in whole O-D regions X-Y coordinates (Xc、Yc) be：
In the present embodiment, the X-Y coordinate of the central point in the O-D regions in table 5a and table 5b be respectively (7942.9257,
6161.25437) and (6860.4436,6228.4623)；
Step 3.4, F1 (is denoted as according to the coordinate of first point before and after the central point C and O-D regions in O-D regions
And B1), F1-C is connected with C-B1, it is assumed that F1 coordinates are (XF, YF), and B1 coordinates are (XB, YB), and C coordinates are (XC, YC),
Know that three point coordinates ask poor (i.e. path) the ∠ C of the F1-C and C-B1 angle of cut, that is, deviation angles of the C-B1 compared to F1-C：
In above formula, DFCFor the distance between node F1 and node C, DCBFor the distance between node C and node B1, DFBFor
The distance between node F1 and node B1.
In the present embodiment, the O-D regions of sample of users trip Time-space serial and its anterior-posterior approach in table 5a and table 5b
Angle of cut difference is respectively 38.75 degree and 175.17 degree；
Step 3.5, the analysis according to Time-space serial of being gone on a journey to sample, what obtained node space cluster and O-D points differentiated
The nodal distance density and node number of parameter, i.e. O-D regions, and user's travel path is in the angle of turning back in O-D points region
Poor size, average value processing is carried out to the analysis result of great amount of samples data, obtains the criterion of batch O-D points identification, i.e. O-
The threshold limit value Thr- ρ and node number threshold limit value Thr- ε of the nodal distance density in D regions, and it is being unsatisfactory for Thr- ρ
In the case of, O-D points both ends stroke is turned back differential seat angle Thr- ∠.
In the present embodiment, after great amount of samples is analyzed, the threshold limit value of the nodal distance density in O-D regions is obtained
Thr- ρ, node number threshold limit value Thr- ε are that 3, O-D points both ends stroke is turned back differential seat angle Thr- ∠.When nodal distance density
More than when 0.015, can directly be determined as O-D points, between 0.007 and Thr- ρ then need judge Thr- ∠, as Thr- ∠
During more than 120 degree, it is determined as O-D points；
Step 4, based on DBSCAN clustering algorithms, design a kind of Spatial Clustering based on distance, obtained in step 2
Clustering parameter on the basis of, to sample trip Time-space serial data on interpolation point carry out space clustering, extract egress
Clustering cluster (Cluster), therefrom extract the most crucial point (CorePt) in clustering cluster；Calculate from most crucial point, to section
The deflection of first node before and after point clustering cluster, calculate the angle of cut (angle of trip Time-space serial of the user before and after node cluster
Degree is poor)；
Step 4.1, read what is handled by step 2 from database, currently pending user trip Time-space serial
Data, begun stepping through from start time, centered on each node, search the Thr- ε neighbor nodes in its front and rear neighborhood
It is (front and rear eachIt is individual)；
The present embodiment is in this step with the data instance of table 4, it is known that it is 3 that step 3, which obtains Thr- ε, then the neck of each node
Domain is shown in Table 6.
The central point of table 6 and its neighborhood
Step 4.2, hypothesis currently traverse node n2, then the border of its front and rear point of proximity is nodeWithCount from nodeTo nodeSegmentation distance plus andWherein,According to step 3.2, calculate under node number threshold limit value Thr- ε, the neighborhood centered on node n2
Node density；
In the present embodiment, each neighborhood of a point node density is shown in Table 7.
The node density of 7 each neighborhood of table
Step 4.3, judge whether the node density of the neighborhood centered on point n2 is more than threshold limit value Thr- ρ；
It is reachable for density in the node field if being more than Thr- ρ, the neighborhood is labeled as a quasi- node clustering cluster,
Make its density be connected, its internal all node is all marked into the cluster node that is defined, records a node before and after its outside
Space coordinates；
If being less than Thr- ρ, give up the neighbor domain of node, continue to travel through next node；
In the present embodiment, the neighborhood for meeting Thr- ρ is shown in Table 8.
The node density of table 8 is more than the neighborhood of judgment threshold
| Centre of neighbourhood point | Node density |
| INS16 | 0.572378 |
| INS17 | 2.000000 |
| INS18 | 2.000000 |
| INS19 | 2.000000 |
| INS20 | 2.000000 |
| INS21 | 0.015278 |
| INS22 | 0.007668 |
| INS33 | 0.009140 |
| INS34 | 2.000000 |
| INS35 | 0.009587 |
Step 4.4, when traveled through user go on a journey Time-space serial after, travel through the quasi- node clustering each judged from the beginning
Cluster, judge whether there is common factor between the node that adjacent quasi- node clustering is included.Occur simultaneously if existing, claim this two quasi- sections
Density is reachable between point clustering cluster, and the two quasi- node clustering clusters are merged into one, make its density be connected, recalculate this
The node density of quasi- node clustering cluster after merging；
If new node density is more than Thr- ρ, confirms to merge the two quasi- node clustering clusters, recalculate its node
Density, and record the space coordinates of first node before and after the quasi- node clustering cluster outside newly merged；
If new node density is less than Thr- ρ, the quasi- node clustering cluster of merging is taken apart again, and meet that node is close
Spend under conditions of Thr- ρ and node number Thr- ε, abandoned one by one in the presence of the node to occur simultaneously in former quasi- node clustering cluster.
In the present embodiment, the neighbour put centered on INS16, INS17, INS18, INS19, INS20, INS21, INS22
Domain can merge, and neighborhood of a point can merge centered on INS33, INS34, INS35；Quasi- node clustering cluster 1 after merging wraps
Containing node (INS15~INS23), it is 0.0189 to merge deutomerite dot density, quasi- node clustering cluster 2 include node (INS32~
INS36), it is 0.0094 to merge deutomerite dot density；
The each quasi- node clustering cluster of step 4.5, traversal, the central point of each quasi- node clustering cluster is calculated according to step 3.3
C, after central point C is obtained, the X-Y of first point is sat before and after each quasi- node clustering cluster outside obtained according to step 4.4
Mark, calculate user gone on a journey before and after C points track the angle of cut it is poor, and recorded；
In the present embodiment, the central point of quasi- node clustering cluster 1 is (7831.5091,6533.7194), central point and its
Outside user's trace angle of cut difference is 16.9596 degree, the central point of quasi- node clustering cluster 2 be (9625.5823,
4998.0361), the central point user trace angle of cut difference outside with it is 135.3504 degree；
Step 5, the decision rule obtained according to step 2, differentiate whether the node clustering cluster that clustering algorithm obtains is user
The O-D points gone on a journey on Time-space serial, the obtained O-D clustering clusters of differentiation are arranged, using its most crucial node as O-D points,
The measuring point of its original close positions is searched, its position is projected on map, records its actual position；According to
Extract the trip data of obtained O-D points segmentation user, user's trip track being segmented, having shown starting point.
Quasi- node clustering cluster in every step 5.1, traversal user's trip Time-space serial, calculates what is obtained according to step 4
The angle of cut of node density and front and rear node is poor, according to step 3 obtain decision rule, judge its whether be user trip when
O-D points on empty sequence, if it is determined that being O-D points, then the cluster is labeled as node clustering cluster；
In the present embodiment, the decision rule obtained according to step 3, the node density of quasi- node clustering cluster 1 are more than
0.015, thus be directly determined as node clustering cluster, the node density of quasi- node clustering cluster 2 is 0.0094, between Thr- ρ and
Between 0.007, its central point user trace angle of cut difference outside with it is 135.3504 degree, more than Thr- ∠, therefore is equally sentenced
It is set to node clustering cluster；
Step 5.2, the node clustering cluster identified is marked in user goes on a journey Time-space serial, give up the institute in clustering cluster
There is node, the central point C obtained with step 4.5 is replaced, as O-D points；
In the present embodiment, user's trip Time-space serial after replacement is shown in Table 9.
Table 9 replaces user's trip Time-space serial after central point
Step 5.3, by all O-D spot projections to map, obtain O-D points on map using space correlation method
Real space location name；
In the present embodiment, the entitled A restaurants of the physical location of node clustering cluster 1, the physical location of node clustering cluster 2
Entitled B ultrasound city；
Step 5.4, user is split as end points using O-D gone on a journey Time-space serial, be divided into segment, acquisition divides starting point
User go on a journey track；
In the present embodiment, user's trip Time-space serial after segmentation is as follows：
User's trip Time-space serial after table 10a segmentations
User's trip Time-space serial after table 10b segmentations
User's trip Time-space serial after table 10c segmentations
The present invention is handled and screened for mobile terminal big data, by individual hold mobile terminal and sensor it
Between communications records construct individual trip Time-space serial data, pass through the user that mathematical interpolation completion time interval is unified
Go on a journey Time-space serial data, by the travel behaviour of user with largely in short-term away from space nodes represent, and by great amount of samples
The analysis of trip Time-space serial, obtain rule and design parameter that O-D points differentiate.On this basis, the present invention devises one
Node space clustering method based on DBSCAN algorithms, the reachable individual node of space density is connected with quasi- node clustering cluster,
Its density is set to be connected, so as to cluster the paragraph that egress largely gathers in Time-space serial of going on a journey, and be averaging using weighting
Method calculates each differential seat angle between quasi- node clustering cluster and its front and rear node, passes through quasi- node clustering cluster internal segment dot density
Differentiate whether the quasi- node clustering cluster is O-D nodes with differential seat angle, and Time-space serial of being gone on a journey to user is split.
It is described above, only presently preferred embodiments of the present invention, it is not any to the present invention in form and substantial limit
System, it is noted that, can also be with the premise of the inventive method is not departed from for those skilled in the art
Some improvement and supplement are made, these are improved and supplement also should be regarded as protection scope of the present invention.It is all to be familiar with this professional technology
Personnel, without departing from the spirit and scope of the present invention, made when using disclosed above technology contents
The equivalent variations of a little variation, modification and evolution, it is the equivalent embodiment of the present invention；Meanwhile all essence according to the present invention
The variation, modification and evolution for any equivalent variations that technology is made to above-described embodiment, still fall within technical scheme
In the range of.
Claims (7)
1. under a kind of big data environment travelling OD node identification and hop extracting method, it is characterised in that this method by
5 step compositions below：
Step 1, read the anonymous encryption mobile terminal sensing data obtained from sensor operator, anonymity encryption mobile terminal
Sensing data is continuous in the time and space in theory, and different mobile terminal corresponds to different user unique number EPID, carries
The communication signaling record that each EPID is at the appointed time triggered in section is taken, forms the trip data collection of the EPID；
Step 2, communications records of each EPID at the appointed time in section with sensor are extracted successively, sort, build in chronological order
Vertical user's trip track data collection；From start time t0, the locus of trip data is inserted using T time as interval
Value, build the user's trip Time-space serial being made up of true point and interpolation point；
Minimum node quantity in step 3, Time-space serial of being gone on a journey to user in the node clustering density p and O-D point radiuses of O-D points
ε, and the angle of cut of the front and rear trip Time-space serial of O-D points are analyzed, the initialization of arrange parameter, while formulate sentencing for O-D points
Not rule；
Step 4, based on DBSCAN algorithms, design a kind of Spatial Clustering based on distance；The cluster ginseng obtained in step 2
On the basis of number, space clustering is carried out to the interpolation point that user goes on a journey in Time-space serial data, extracts the clustering cluster of egress, from
In extract most crucial point in clustering cluster；Calculate from most crucial point, to node clustering cluster before and after first node
Deflection, calculate the angle of cut of trip Time-space serial of the user before and after node cluster；
Whether step 5, the decision rule obtained according to step 2, the clustering cluster that discriminating step 4 obtains are user's trip Time-space serials
On O-D points, the obtained O-D clustering clusters of differentiation are arranged, using its most crucial node as O-D points, it is original to search its
The measuring point of close positions, its position is projected on map, record its actual position；The O-D obtained according to extraction
The trip data of point segmentation user, user's trip track being segmented, having shown starting point.
2. the identification of travelling OD node and hop extracting method under a kind of big data environment as claimed in claim 1, its
It is characterised by, the step 1 includes：
Step 1.1, read from the anonymous encryption mobile terminal sensing data of sensor operator acquisition, in theory anonymous encryption shifting
Dynamic terminal sensor data all should be continuous in the time and space, including：User's unique number EPID, communication operation class
Moment TIME, great Qu REGIONCODE, sensing implement body numbering SENSORID residing for sensor occur for type TYPE, communication operation；
Wherein, great Qu REGIONCODE residing for sensor and sensing implement body numbering SENSORID constitute sensor number；
Step 1.2, an anonymous encryption mobile terminal sensing data are that a signaling records, and every signaling record is solved
It is close；
Step 1.3, according to user unique number EPID, inquire about its at the appointed time log all in section, build user
Trip track data.
3. the identification of travelling OD node and hop extracting method under a kind of big data environment as claimed in claim 2, its
It is characterised by, the step 2 includes：
All fixation sensor numbers in user's trip track data that step 2.1, extraction step 1.3 obtain
Latitude and longitude coordinates are converted to geographical coordinate X-Y by REGIONCODE-SENSORID and its corresponding latitude and longitude coordinates LON-LAT；
Step 2.2, traverse user trip track data, it is arranged by triggering call duration time TIMESTAMP orders；
Step 2.3, trip data being begun stepping through from start time, adjacent every 3 communications records point is fitted a conic section,
The x-axis of conic section is the time of user's trip track, and y-axis is the X-Y coordinate of communications records point；If the trip track bag of user
Containing n communications records point, n is positive integer, then needs to fit 2n-4 bar conic sections altogether；
Step 2.4, from integer start time t0, T calculates user in the X-Y coordinate at each time point, phase at timed intervals
With time X (t0+nT) and Y, ((t0+nT) forms an interpolation point, in addition to two sections of head and the tail, all there is 2 between remaining communication point
Matched curve, the X-Y coordinate of interpolation point between are averaging to obtain by the result of calculation of two curves；
Step 2.5, all interpolation points sort in chronological order, and all interpolation points form user's trip Time-space serial.
4. the identification of travelling OD node and hop extracting method under a kind of big data environment as claimed in claim 1, its
It is characterised by, the step 3 includes：
Step 3.1, the user's trip Time-space serial for choosing O-D points in some clear and definite traces form sample space-time as sample
Sequence, analyze and mark its O-D region, O-D regions are O-D node sets；
Step 3.2, traversal sample Time-space serial, find out first node and last node in each O-D regions, it is assumed that
There is n1 node, n1 is positive integer, and since first node, segmentation calculates the space length between O-D Area Nodes, altogether
There are n1-1 sections, these distances are summed up, calculating O-D Area Node density ps is
Step 3.3, the weighted center for calculating O-D regions, weight w are sequence number of the O-D regional nodes apart from its central point that sorts
Difference, if O-D regional nodes number is n1, the weight w minimums of the 1st and the n-th 1 node, theThe power of individual node
Weight is maximum, the central point C in whole O-D regions X-Y coordinate (Xc、Yc) be：
Step 3.4, the coordinate according to first point F1, B1 before and after the central point C and O-D regions in O-D regions, by F1-C and C-
B1 is connected, it is assumed that F1 coordinates are (XF, YF), and B1 coordinates are (XB, YB), and C coordinates are (XC, YC), it is known that three point coordinates seek F1-C
With C-B1 angle of cut difference ∠ C：
Wherein, DFCFor the distance between node F1 and node C, DCBFor the distance between node C and node B1, DFBFor node F1
The distance between node B1；
Step 3.5, the analysis according to Time-space serial of being gone on a journey to sample, the parameter that obtained node space cluster and O-D points differentiates,
That is the nodal distance density and node number in O-D regions, and turn back differential seat angle of user's travel path in O-D points region are big
It is small, average value processing is carried out to the analysis result of sample data, obtains the section of the criterion, i.e. O-D regions of the identification of batch O-D points
The threshold limit value Thr- ρ and node number threshold limit value Thr- ε of point Distance Density, and in the case where being unsatisfactory for Thr- ρ, O-
D points both ends stroke is turned back differential seat angle Thr- ∠.
5. the identification of travelling OD node and hop extracting method under a kind of big data environment as claimed in claim 4, its
It is characterised by, in the step 3.2, the calculation formula of O-D Area Node density psIn, it is to ensure road that molecule, which subtracts 1,
Hop count is consistent with nodes, denominator add 1 be prevent 0 situation.
6. the identification of travelling OD node and hop extracting method under a kind of big data environment as described in claim 4 or 5,
Characterized in that, the step 4 includes：
Step 4.1, the currently pending user's trip Time-space serial data handled by step 2 are read from database,
Begun stepping through from start time, centered on each node, search the Thr- ε neighbor nodes in its front and rear neighborhood, front and rear neighbour
Domain is eachIt is individual；
Step 4.2, assume that it is positive integer currently to traverse node n2, n2, then the border of its front and rear point of proximity is nodeWithCount from nodeTo nodeSegmentation distance plus andWherein,According to step 3.2, calculate under node number threshold limit value Thr- ε, using node n2 in
The node density of the neighborhood of the heart；
Step 4.3, judge whether the node density of the neighborhood centered on point n2 is more than threshold limit value Thr- ρ；
It is reachable for density in the node field if being more than Thr- ρ, the neighborhood is labeled as a quasi- node clustering cluster, makes it
Density is connected, and its internal all node is all marked into the cluster node that is defined, and records the sky of a node before and after its outside
Between coordinate；
If being less than Thr- ρ, give up the neighbor domain of node, continue to travel through next node；
Step 4.4, when traveled through user go on a journey Time-space serial after, travel through the quasi- node clustering cluster each judged from the beginning, sentence
Whether there is common factor between the node that disconnected adjacent quasi- node clustering is included；Occur simultaneously if existing, claim two quasi- node clusterings
Density is reachable between cluster, and the two quasi- node clustering clusters are merged into one, make its density be connected, after recalculating the merging
The node density of quasi- node clustering cluster；
If new node density is more than Thr- ρ, confirms to merge the two quasi- node clustering clusters, record the standard newly merged again
The space coordinates of first node before and after outside node clustering cluster；
If new node density is less than Thr- ρ, the quasi- node clustering cluster of merging is taken apart again, and meet node density Thr-
Under conditions of ρ and node number Thr- ε, abandoned one by one in the presence of the node to occur simultaneously in former quasi- node clustering cluster；
The each quasi- node clustering cluster of step 4.5, traversal, the central point C of each quasi- node clustering cluster is calculated according to step 3.3,
After obtaining central point C, the X-Y coordinate of first point, is calculated before and after each quasi- node clustering cluster outside obtained according to step 4.4
User gone on a journey before and after C points track the angle of cut it is poor, and recorded.
7. the identification of travelling OD node and hop extracting method under a kind of big data environment as claimed in claim 6, its
It is characterised by, the step 5 includes：
Quasi- node clustering cluster in every step 5.1, traversal user's trip Time-space serial, the node obtained is calculated according to step 4
The angle of cut of density and front and rear node is poor, the decision rule obtained according to step 3, judges whether it is user's trip Time-space serial
On O-D points, if it is determined that being O-D points, then the cluster is labeled as node clustering cluster；
Step 5.2, the node clustering cluster identified is marked in user goes on a journey Time-space serial, give up all sections in clustering cluster
Point, the central point C obtained with step 4.5 is replaced, as O-D points；
Step 5.3, by all O-D spot projections to map, use space correlation method to obtain reality of the O-D points on map
Locus title；
Step 5.4, using O-D as end points split user go on a journey Time-space serial, obtain divide starting point user go on a journey track.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710843841.2A CN107770744A (en) | 2017-09-18 | 2017-09-18 | The identification of travelling OD node and hop extracting method under big data environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710843841.2A CN107770744A (en) | 2017-09-18 | 2017-09-18 | The identification of travelling OD node and hop extracting method under big data environment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107770744A true CN107770744A (en) | 2018-03-06 |
Family
ID=61265490
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710843841.2A Pending CN107770744A (en) | 2017-09-18 | 2017-09-18 | The identification of travelling OD node and hop extracting method under big data environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107770744A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109104694A (en) * | 2018-06-26 | 2018-12-28 | 重庆市交通规划研究院 | A kind of user stop place discovery method and system based on mobile phone signaling |
| CN109348404A (en) * | 2018-10-09 | 2019-02-15 | 上海世脉信息科技有限公司 | A kind of method that individual trip path locus extracts under big data environment |
| CN109613553A (en) * | 2018-12-18 | 2019-04-12 | 歌尔股份有限公司 | The method, apparatus and system of physical quantities in scene are determined based on laser radar |
| CN109903561A (en) * | 2019-03-14 | 2019-06-18 | 智慧足迹数据科技有限公司 | Flow of the people calculation method, device and electronic equipment between section |
| CN110083599A (en) * | 2019-03-28 | 2019-08-02 | 华东师范大学 | A kind of track of vehicle data index method based on temporal-spatial interpolating |
| CN110232421A (en) * | 2019-06-21 | 2019-09-13 | 福州大学 | One kind merging OD step by step and flows to space-time joint clustering method |
| CN110489530A (en) * | 2018-05-10 | 2019-11-22 | 上海申通地铁集团有限公司 | Similar station for acquiring method and system based on word2vec |
| CN111046895A (en) * | 2018-10-15 | 2020-04-21 | 北京京东尚科信息技术有限公司 | Method and device for determining target area |
| CN111274475A (en) * | 2020-01-16 | 2020-06-12 | 深圳市前海随手数据服务有限公司 | Method and device for extracting track features, storage medium and terminal |
| CN111366160A (en) * | 2020-05-25 | 2020-07-03 | 深圳市城市交通规划设计研究中心股份有限公司 | Path planning method, path planning device and terminal equipment |
| CN111625610A (en) * | 2020-05-25 | 2020-09-04 | 重庆大学 | Display method and system for generating track based on positioning points |
| CN112733112A (en) * | 2020-12-31 | 2021-04-30 | 恒安嘉新（北京）科技股份公司 | User travel mode determining method and device, electronic equipment and storage medium |
| CN115297441A (en) * | 2022-09-30 | 2022-11-04 | 上海世脉信息科技有限公司 | Method for calculating robustness of individual space-time activity in big data environment |
| CN112733112B (en) * | 2020-12-31 | 2024-05-03 | 恒安嘉新（北京）科技股份公司 | Method and device for determining travel mode of user, electronic equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7376957B1 (en) * | 2002-12-16 | 2008-05-20 | At&T Delaware Intellectual Property, Inc. | Method and system for recovering stranded outbound messages |
| CN105740904A (en) * | 2016-01-29 | 2016-07-06 | 东南大学 | Travel and activity mode identification method based on DBSCAN clustering algorithm |
| CN105809962A (en) * | 2016-06-13 | 2016-07-27 | 中南大学 | Traffic trip mode splitting method based on mobile phone data |
| CN106067154A (en) * | 2016-05-30 | 2016-11-02 | 上海华企软件有限公司 | A kind of intercity migration passenger flow analysing method based on the big data of mobile phone |
| CN106096631A (en) * | 2016-06-02 | 2016-11-09 | 上海世脉信息科技有限公司 | A kind of recurrent population's Classification and Identification based on the big data of mobile phone analyze method |
| US20160334232A1 (en) * | 2015-05-11 | 2016-11-17 | General Motors, Llc | Real-time carpooling |
| CN106781463A (en) * | 2016-12-02 | 2017-05-31 | 北京中创信测科技股份有限公司 | A kind of method that urban road flow velocity is calculated based on mobile phone signaling and OD attributes |
-
2017
- 2017-09-18 CN CN201710843841.2A patent/CN107770744A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7376957B1 (en) * | 2002-12-16 | 2008-05-20 | At&T Delaware Intellectual Property, Inc. | Method and system for recovering stranded outbound messages |
| US20160334232A1 (en) * | 2015-05-11 | 2016-11-17 | General Motors, Llc | Real-time carpooling |
| CN105740904A (en) * | 2016-01-29 | 2016-07-06 | 东南大学 | Travel and activity mode identification method based on DBSCAN clustering algorithm |
| CN106067154A (en) * | 2016-05-30 | 2016-11-02 | 上海华企软件有限公司 | A kind of intercity migration passenger flow analysing method based on the big data of mobile phone |
| CN106096631A (en) * | 2016-06-02 | 2016-11-09 | 上海世脉信息科技有限公司 | A kind of recurrent population's Classification and Identification based on the big data of mobile phone analyze method |
| CN105809962A (en) * | 2016-06-13 | 2016-07-27 | 中南大学 | Traffic trip mode splitting method based on mobile phone data |
| CN106781463A (en) * | 2016-12-02 | 2017-05-31 | 北京中创信测科技股份有限公司 | A kind of method that urban road flow velocity is calculated based on mobile phone signaling and OD attributes |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110489530A (en) * | 2018-05-10 | 2019-11-22 | 上海申通地铁集团有限公司 | Similar station for acquiring method and system based on word2vec |
| CN109104694A (en) * | 2018-06-26 | 2018-12-28 | 重庆市交通规划研究院 | A kind of user stop place discovery method and system based on mobile phone signaling |
| CN109104694B (en) * | 2018-06-26 | 2020-10-30 | 重庆市交通规划研究院 | User stay position finding method and system based on mobile phone signaling |
| CN109348404B (en) * | 2018-10-09 | 2020-10-09 | 上海世脉信息科技有限公司 | Method for extracting individual travel road track in big data environment |
| CN109348404A (en) * | 2018-10-09 | 2019-02-15 | 上海世脉信息科技有限公司 | A kind of method that individual trip path locus extracts under big data environment |
| CN111046895B (en) * | 2018-10-15 | 2023-11-07 | 北京京东振世信息技术有限公司 | Method and device for determining target area |
| CN111046895A (en) * | 2018-10-15 | 2020-04-21 | 北京京东尚科信息技术有限公司 | Method and device for determining target area |
| CN109613553A (en) * | 2018-12-18 | 2019-04-12 | 歌尔股份有限公司 | The method, apparatus and system of physical quantities in scene are determined based on laser radar |
| CN109903561A (en) * | 2019-03-14 | 2019-06-18 | 智慧足迹数据科技有限公司 | Flow of the people calculation method, device and electronic equipment between section |
| CN110083599A (en) * | 2019-03-28 | 2019-08-02 | 华东师范大学 | A kind of track of vehicle data index method based on temporal-spatial interpolating |
| CN110083599B (en) * | 2019-03-28 | 2023-06-23 | 华东师范大学 | Vehicle track data indexing method based on space-time interpolation |
| CN110232421B (en) * | 2019-06-21 | 2022-05-24 | 福州大学 | Step-by-step combined OD flow direction space-time combined clustering method |
| CN110232421A (en) * | 2019-06-21 | 2019-09-13 | 福州大学 | One kind merging OD step by step and flows to space-time joint clustering method |
| CN111274475A (en) * | 2020-01-16 | 2020-06-12 | 深圳市前海随手数据服务有限公司 | Method and device for extracting track features, storage medium and terminal |
| CN111366160B (en) * | 2020-05-25 | 2020-10-27 | 深圳市城市交通规划设计研究中心股份有限公司 | Path planning method, path planning device and terminal equipment |
| CN111625610A (en) * | 2020-05-25 | 2020-09-04 | 重庆大学 | Display method and system for generating track based on positioning points |
| CN111366160A (en) * | 2020-05-25 | 2020-07-03 | 深圳市城市交通规划设计研究中心股份有限公司 | Path planning method, path planning device and terminal equipment |
| CN112733112A (en) * | 2020-12-31 | 2021-04-30 | 恒安嘉新（北京）科技股份公司 | User travel mode determining method and device, electronic equipment and storage medium |
| CN112733112B (en) * | 2020-12-31 | 2024-05-03 | 恒安嘉新（北京）科技股份公司 | Method and device for determining travel mode of user, electronic equipment and storage medium |
| CN115297441A (en) * | 2022-09-30 | 2022-11-04 | 上海世脉信息科技有限公司 | Method for calculating robustness of individual space-time activity in big data environment |
| CN115297441B (en) * | 2022-09-30 | 2023-01-17 | 上海世脉信息科技有限公司 | Method for calculating robustness of individual space-time activity in big data environment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107770744A (en) | The identification of travelling OD node and hop extracting method under big data environment | |
| Bachir et al. | Inferring dynamic origin-destination flows by transport mode using mobile phone data | |
| JP5871566B2 (en) | System and method for population tracking, counting and movement estimation using mobile motion data and / or geographic information in a mobile network | |
| Asgari et al. | A survey on human mobility and its applications | |
| CN102521973B (en) | A kind of mobile phone switches the road matching method of location | |
| WO2016086786A1 (en) | Geo-fencing generation method and apparatus | |
| CN108320501A (en) | Public bus network recognition methods based on user mobile phone signaling | |
| WO2017107800A1 (en) | Method of acquiring route hotspot of traffic road and device | |
| Fiadino et al. | Steps towards the extraction of vehicular mobility patterns from 3G signaling data | |
| CN107977673A (en) | A kind of economically active population's recognition methods based on big data | |
| Kumar et al. | Understanding urban mobility via taxi trip clustering | |
| Leung et al. | Intra-city urban network and traffic flow analysis from GPS mobility trace | |
| CN106651027A (en) | Internet regular bus route optimization method based on social network | |
| CN109684373A (en) | Emphasis party based on trip and call bill data analysis has found method | |
| CN109348404B (en) | Method for extracting individual travel road track in big data environment | |
| Coscia et al. | Optimal spatial resolution for the analysis of human mobility | |
| CN106254123A (en) | A kind of towards the mapping method of network topology in Metropolitan Area Network (MAN) rank AS territory | |
| Cao et al. | Understanding metropolitan crowd mobility via mobile cellular accessing data | |
| CN111291145B (en) | Mapping method, device and storage medium of wireless hotspot and interest point | |
| CN108573265B (en) | People flow statistical method and statistical system | |
| CN106157601B (en) | A kind of investigation method of the bus passenger flow demand based on mobile data | |
| Danafar et al. | Bayesian framework for mobility pattern discovery using mobile network events | |
| CN112381078A (en) | Elevated-based road identification method, elevated-based road identification device, computer equipment and storage medium | |
| Fan et al. | Discovering urban social functional regions using taxi trajectories | |
| CN111970685B (en) | One-person multi-card identification method in big data environment |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180306 |