CN106067154A - A kind of intercity migration passenger flow analysing method based on the big data of mobile phone - Google Patents
A kind of intercity migration passenger flow analysing method based on the big data of mobile phone Download PDFInfo
- Publication number
- CN106067154A CN106067154A CN201610367737.6A CN201610367737A CN106067154A CN 106067154 A CN106067154 A CN 106067154A CN 201610367737 A CN201610367737 A CN 201610367737A CN 106067154 A CN106067154 A CN 106067154A
- Authority
- CN
- China
- Prior art keywords
- node
- section
- time
- speed
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- 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
Abstract
The invention provides a kind of intercity migration passenger flow analysing method based on the big data of mobile phone.The present invention is big to magnanimity mobile phone, and data process and screen, build the Time-space serial data of individuality trip, combine decision rule by cluster analysis the function of the trip mode in individual migration flow process and each city is judged, analyze general layout and the trend of the flowing of population interzone, and each city institute's role wherein.
Description
Technical field
The present invention relates to a kind of population based on magnanimity anonymity encryption mobile phone individuality time series location data trans-city move
Behavior analysis method is moved in advection, builds individuality trip Time-space serial data according to individual time and spatial position data, arranges
Decision rule judges that it goes out start of line city-destination city, and the space mobile sequence that individuality is long-term is carried out cutting, to it
Gait of march clusters, it determines the intermediate node city in trip route and transit point city, sentences its trip mode
Disconnected.The present invention may be used for obtaining population in intercity migration flowing law and present characteristics, and each city is moved at population
Status and effect during shifting, for traffic policy assessment, traffic programme, and multizone population, resource and environmental management and association
The exhibition of readjusting the distribution provides service, belongs to the technical field of region decision rule and economic management.
Background technology
Inter-urban migration flowing be the important behavior in national economy activity, always by economics, demography,
Neo-Confucianism, Resources and environmental sciences, regional planning, and the attention of relevant multiple cross discipline, gone through centuries discussion and
Research, has derived substantial amounts of model, method, has created a lot of epoch-making conclusion.But still suffer from many weak points.
Mainly rise the development in classical economics about the trans-regional scientific research migrating flowing of population in early days, thus spread out
Bear the mathematical model in a large number about population migration.But owing to calculating the deficiency of instrument, the structure of these models is the most all with certainly
Upper and under be main (up-down), mostly have the strongest external hypothesis (as maximization of welfare assume, labor rate equilibrium vacation
If etc.), often produce bigger discrepancy with reality, cause result tool theory significance on foot, and more difficult meet with reality.
Along with greatly developing of computer technology, occur in that and give full play to the computable of computer high-speed computational capability in a large number
Bottom-up modeling (bottom-up).Especially object-oriented simulates (object-oriented simulation) method
Emerge so that population migration is studied to the epoch entering microcomputer modelling.In recent years, along with microcosmic Simulation (micro
Simulation), based on Agent simulation (agent-based simulation) method development, to population migration flowing
Research starts to refine to the modeling to individual molecular behavior pattern.Now, in early days limited model development mathematical tool and calculating
Ability the most no longer becomes main restricting factor, and the disappearance of data becomes the yoke of model development.
It is traditional that to research and analyse, for migration between demographic region, the regional population being based primarily upon in census data clean
Move into data.But this data statistics yardstick is relatively thick, and the net change value of only population, for the migratory direction that population is concrete
With migration path no record.Additionally, the most with good grounds civil aviation authority and Ministry of Communications are given in statistical yearbook or other Data Sources
Urban Passenger Traffic freight volume carry out estimating.These data equally exist the problem that statistical yardstick is thicker, simultaneously its principal statistical
Civil aviaton and railway transportation main services far way trip, and cannot add up short distance self-driving trip.
In recent years, along with the development of information technology, data message amount presents explosive growth, and for population migration and stream
Dynamic behavior, its Data Source gets more and more, and data volume is more and more huger.The present invention is based on the mobile phone intercity people of big data analysis
Family migrates flow behavior, will migrate the most careful person of actually occurring to transfer behavior of object of study of flow behavior itself.Extremely
2015, cellphone subscriber reached 13.06 hundred million, accounted for more than the 96% of total population, the signal letter that mobile phone terminal equipment persistently produces
Breath, defines the volume of data collection of record user's trip, provides Data Source for population migration flow analysis.But it is obtainable
Data in mobile phone quality is uneven, and this is accomplished by big data are carried out data mining and process.
Summary of the invention
It is an object of the invention to the space operation data set in the range of utilizing mobile terminal individuality at the appointed time, excavate big
The trip Time-space serial data that amount is individual, analyze individual trip mode and each city role in migrating flowing activity, and
These behaviors are carried out the statistics of space-time unique, thus objective ground reflect population at intercity migration flow structure and
Trend.
In order to achieve the above object, the technical scheme is that and provide a kind of intercity moving based on the big data of mobile phone
Move passenger flow analysing method, it is characterised in that comprise the following steps:
Step 1, from common carrier obtain different cities continuous print anonymity on the time and space encryption data in mobile phone,
For each EPID, it is calculated each data in mobile phone in the appointment time period according to the anonymous encryption data in mobile phone of current EPID
Corresponding true longitude coordinate LON and true latitude coordinate LAT, is defined as different region LOC, according to meter by different cities
The all of true longitude coordinate LON obtained and true latitude coordinate LAT, it is thus achieved that residing in current EPID at the appointed time section
The number of region LOC, the number rejecting region LOC is not more than all EPID of 1, then obtain at the appointed time all in section
Trans-city EPID, and the EPID that these are trans-city is recorded into target database；
Step 2, according to the whole communications records in EPID at the appointed time section each in target database, with region LOC
For space dividing unit, the time that current EPID communication activity in each region LOC occurs is ranked up, is formed current
EPID is the time shaft of communication behavior in each region LOC；On this basis, the true warp in being recorded by each of time shaft
Degree coordinate LON and true latitude coordinate LAT, carries out geographical mapping, projects geographical space according to its latitude and longitude coordinates, generate
Space-time data；Utilize geographical adjacent_lattice that space-time data is carried out spatial continuity inspection；Calculate each time in space-time data
Length in duration and this section of path theory of section, calculates user's gait of march at this section with this, it is thus achieved that user goes on a journey space-time
Sequence set, comprises the following steps
Step 2.1, traversal have been stored in the EPID meeting trans-city trip condition of target database, in target database
Searching all records of current EPID in specifying the time period, the corresponding information in recording every mates with geodata,
Falling in actual map using every record as a data point, each data point at least includes: EPID, record time of origin
TIME, true longitude coordinate LON, true latitude coordinate LAT, region LOC；
Step 2.2, in units of the LOC of region, search in current EPID at the appointed time section inner segment in regional LOC
The data point corresponding to Article 1 communications records FirstRec being recorded is as a node and the last item communications records
Data point corresponding to LastRec is as a node, and enters, according to the time of origin of every communications records, the node obtained
Row sequence, builds section data, and the field in the data of section at least includes: section beginning node LNODE, section terminal note NNODE,
The time TIME spent through this section, section data and node data are collectively forming full-time null sequence data；
Step 2.3, travel through full-time null sequence data, labelling departure place therein-destination's node；
Step 2.4, the full-time null sequence data of current EPID is carried out cutting according to departure place-destination's node, formed
A plurality of only head and the tail node is the single travel time sequence of departure place-destination's node；
Step 2.5, combine inter-city railway network and Class III highway network data or intercity adjacency matrix extracts section data
Air line distance, highway distance and rail distance, be path DISTANCE；
Step 2.6, utilizing path DISTANCE and time TIME, be calculated each section carries out speed
SPEED, and speed SPEED will be carried out add in single travel time sequence, when step 2.5 uses inter-city railway network and
During Class III highway network data, it is thus achieved that three kinds of speed: space rate SPEED-S, high ferro speed SPEED-H and network of highways speed
SPEED-R；When step 2.5 uses intercity adjacency matrix, it is thus achieved that two kinds of speed: space rate SPEED-S, adjacent
Matrix speed SPEED-M；
Step 3, according in the single travel time sequence of current EPID carry out speed SPEED differentiate non-departure place-mesh
The node type of ground node, and finally determine the trip mode of current EPID with this, comprise the following steps:
Step 3.1, first speed SPEED that carries out in each section in the single travel time sequence of current EPID is carried out
Cluster, selects the significant lower section of speed as velocity anomaly section；
Step 3.2, according to the first node LNODE of velocity anomaly section and tail node NNODE, meet one of following condition, then
City, place, velocity anomaly section is judged as intermediate node:
If the first node LNODE in condition one velocity anomaly section and tail node NNODE are in same city, then this city of labelling
City is intermediate node；
If the first node LNODE in condition two velocity anomaly section and tail node NNODE be not in same city, then foundation
Velocity anomaly section, in first node LNODE and the length in tail node NNODE city, takes city corresponding to distance elder as centre
Node；
If how velocity anomaly section across, its first node LNODE and tail node NNODE do not adjoin, then without
Method differentiates that wherein which node is intermediate node, abandons mark or abandons current single travel time sequence；
Step 3.3, by remaining node in addition to departure place-destination's node and intermediate node in single travel time sequence
All it is labeled as transit point；
Step 3.4, extract the speed in section between O-D node and intermediate node, compare the average speed of each trip mode,
Differentiate the trip mode used of going on a journey each time；
The single travel time sequence data of a large amount of EPID that step 4, basis obtain, and the trip mode supporting with it
With nodal community data, analyze the flow of intercity migration passenger flow, the flow direction, and the merit that each city is in artificial abortion's transition process
Energy.
Preferably, described step 1 comprises the following steps:
Step 1.1, obtaining anonymous data in mobile phone in real time from common carrier, anonymous data in mobile phone includes: EPID, TYPE,
TIME, X, Y, SR, LOC, wherein:
EPID is the unique mobile subscriber identification code in the anonymous One-Way Encryption whole world, is that common carrier carries out list to each user
To irreversible encryption, thus uniquely identify each user, and do not expose Subscriber Number privacy information；
TYPE is current anonymous network action type involved by data in mobile phone；
TIME is the generation moment of current anonymous network action involved by data in mobile phone；
X, Y are to utilize special coordinate encryption method to be added after the true coordinate position of cellphone subscriber is encrypted calculating
Close coordinate；
SR is spatial dimension, is current anonymous data in mobile phone sterically defined deviations scope；
Step 1.2, to the anonymous data in mobile phone received, utilize special deciphering module to carry out Coordinate Conversion, after deciphering
The content of data in mobile phone includes in real time, and EPID, TYPE, TIME, true longitude coordinate LON, true latitude coordinate LAT, space are by mistake
Difference scope SR；
Step 1.3, based on the true longitude coordinate LON in real-time data in mobile phone and true latitude coordinate LAT, it is judged that current
The city, user place that data in mobile phone is corresponding in real time, gives region LOC attribute, travels through the data in mobile phone record of current EPID, logical
Cross the true longitude coordinate LON in each of time shaft record and true latitude coordinate LAT and carry out geographical mapping, according to longitude and latitude
Coordinate projects geographical space, in conjunction with territory, face, National urban figure layer, it is thus achieved that city residing for current EPID, by this city name assignment
For region LOC attribute；
The EPID of trans-city travel behaviour, tool was there is not in step 1.4 in rejecting at the appointed time section according to region LOC
Body rule is: if an EPID at the appointed time in the range of there is two or more regions LOC, then it represents that this EPID exist
Trans-city travel behaviour, records this EPID, is stored in target database, otherwise abandons this EPID, carries out the differentiation of next user.
Preferably, in described step 2.3, the flag condition of departure place-destination's node is for meeting one of following condition:
Condition one, add up current EPID position, incity, each city scatterplot distribution and Annual distribution, if this EPID is at certain
The time of staying in one city is more than time threshold Stay_1, the most directly regards this city destination as a stroke, and next
The starting point of Duan Hangcheng, is i.e. labeled as departure place-destination's node；
If the current EPID of condition two is in current city time of staying overtime threshold value Stay_2, but less than time threshold
Stay_1, and it is located at least in the joint of the described full-time null sequence data of non-traffic route, station, airport, Expressway Service
Current city more than 50%, is then labeled as departure place-destination's node by some accounting；
If condition three user non-traffic route, station, airport, Expressway Service length value Stay_R straight
Continue time of staying overtime threshold value Stay_3 in region, footpath, then by diameter region, this is judged to departure place-destination's node；
Condition four, current EPID full-time null sequence data in head and the tail node location be accordingly to be regarded as departure place-destination
Node.
Preferably, in described step 3.1, the criterion in velocity anomaly section is:
If cluster result is 1 class, or different classes of between gaps between their growth rates the least, there is not obvious velocity anomaly point,
Then present single travel time sequence is through, abnormal section does not occur；
If cluster result is more than 2 classes or two classes, and different classes of between speed difference very big, then with friction speed classification
Intersection both sides, section, slow section is velocity anomaly section.
Preferably, when cluster result is more than 2 classes or two classes, particular situation can be divided into:
If situation 1 cluster result is 2 classes, and the relatively low section of speed with discrete formal distribution in the single travel time
In sequence, then the section that speed is relatively low is velocity anomaly section；
If situation 2 cluster result is 2 classes, and the relatively low section of speed is a lot, and the most two kinds of section is spatially
Substantially continuous, then judge that, in the boundary section in two class sections, that section slow is velocity anomaly section；
If situation 3 cluster result is more than 2 classes, and the section Discrete Distribution that speed is relatively low, and the classification base that speed is higher
This continuous distribution, then judge that the slowest that section of speed is as velocity anomaly section；
If situation 4 cluster result is more than 2 classes, and each speed category the most all presents continuous distribution, then judge that speed is
That slow section is velocity anomaly section；
If the section of situation 5 friction speed classification is alternately present, there is no obvious rule can follow, then also according to situation 2
Process, i.e. with the slower section of speed, speed separation both sides for velocity anomaly section.
Preferably, in described step 3.4, take each section speed of departure place-between destination's node and intermediate node
The speed gone on a journey as this of median:
The discrimination standard of civil aviaton is for carrying out speed more than 500km/h, and must there is flight between its stroke head and the tail city,
If gait of march does not has flight more than between 500km/h and head and the tail city, then show that current EPID is in other city boarding or fall
Falling, now detection range head node LNODE and the nearest airport of tail node NNODE carry out having mended, if near there is third party city
City to first node LNODE or the flight of tail node NNODE, then adds airport joint between first node LNODE and tail node NNODE
Point, calculate newly search airport to the first node LNODE nearer apart from it or the distance of tail node NNODE, the travel time of this section
Travel time for protocercal tail node LNODE to tail node NNODE deducts the theoretic flight time, then judges first node
The vehicles that LNODE or tail node NNODE are used to airport, if suitable flight cannot be searched to obtain, then abandon this record.
The discrimination standard of high ferro be middle bit rate between 150-500km/h, and the judgement of below 150km/h is highway
Trip.
Preferably, described step 4 is analyzed the flow of intercity migration passenger flow, the flow direction, and each city is at people's stream migration
During function include:
Extracting departure place-destination's node of all previous trip of all EPID, the flow direction of statistics passenger flow migration and flow, with line
The thickness of section represents the size of the volume of the flow of passengers, instruction passenger flow direction, line segment arrow direction；
Travel amount according to the every kind of trip mode in each section and trip directional statistics migrate passenger flow on detailed path
Flow/flow direction, represents the size of the volume of the flow of passengers, instruction passenger flow direction, line segment arrow direction equally with the thickness of line segment；
Adding up in the appearance record of all effective EPID, each city is as departure place-destination's node, intermediate node and mistake
The frequency that border point is occurred, analyzes each city ratio of played the part of different role in passenger flow transition process, analyzes each city and exist
Dominant role in passenger flow transition process.
The invention have the advantage that and leverage fully on the big data resource of existing mobile phone, utilize existing sea in mobile communications network
The encryption position information that the anonymous cellphone subscriber of amount is lasting, can low cost, automatization, easily obtain specify time range in big
Amount population, in the data of intercity distribution, uses movement of population the most direct, the bottom to migrate tracking data analysis population
Transfer behavior, compensate for deficiency and the system using the macro-data of Ministry of Communications and State Statistics Bureau's issue to cover for trip mode
The defects such as meter result is the most coarse.
Accompanying drawing explanation
Fig. 1 is based on big data the long-time migration passenger flow analysing method group method figure that the present invention proposes；
Fig. 2 is wherein data preprocessing module method figure；
Fig. 3 is Data subset, the extraction of O-D point and gait of march computational methods figure；
Fig. 4 is that intermediate node extracts and trip mode discrimination method figure.
Detailed description of the invention
For making the present invention become apparent, hereby with preferred embodiment, and accompanying drawing is coordinated to be described in detail below.
Step 1, from common carrier obtain anonymity encryption data in mobile phone, it is desirable to data are all continuous in the time and space
, to each EPID (the unique mobile subscriber identification code in the anonymous One-Way Encryption whole world, EncryPtion international
Mobile subscriber IDentity) at the appointed time in section T, the communication behavior triggered carries out query processing analysis,
Judge whether this EPID occurred trans-city travel behaviour, inquiry acquisition that trans-city trip occurred in the range of this control time
The EPID of behavior all communications records in the range of control time.
The big data of step 1.1, mobile phone refer to mobile communication operator from mobile communications network, fixed broadband network, nothing
Line WIFI and location-based service be correlated with APP etc. obtain in real time and desensitize encryption after anonymous cellphone subscriber's seasonal effect in time series encrypted bits
Confidence ceases.Obtaining real-time anonymous data in mobile phone in real time from mobile communication operator, the big data content of real-time mobile phone includes: EPID,
TYPE, TIME, X, Y, SR, LOC, see the Chinese patent of Application No. 201610273693.0.It is specifically described as follows:
EPID (the unique mobile subscriber identification code in the anonymous One-Way Encryption whole world, EncryPtion international
Mobile subscriber IDentity), it is that mobile communication operator carries out unidirectional irreversible add to each cellphone subscriber
Close, thus uniquely identify each cellphone subscriber, and do not expose Subscriber Number privacy information, it is desirable to after each cellphone subscriber encryption
EPID keeps the EPID of each cellphone subscriber of uniqueness, i.e. any time keep constant and do not repeat with other cellphone subscriber.
TYPE, is the cell phone network type of action involved by current record, e.g., surf the Net, converse, calling and called, transmitting-receiving note,
Community switching, switching on and shutting down etc..
TIME refers to the cell phone network action involved by current record and the moment occurs, and unit is millisecond.
X, Y, SR are the space encrypted location range information that the cell phone network action involved by current record occurs.X, Y by
Operator utilizes special coordinate encryption method to obtain after the true coordinate position of cellphone subscriber is encrypted calculating, SR (space
Scope, Spatial Range) it is current record sterically defined deviations scope, unit is rice, i.e. current phone user goes out
Point centered by X, Y now, in SR is the spatial dimension of radius.Location algorithm is responsible for by mobile communication operation with positioning precision,
Along with the development of the wireless communication technologys such as 4G, 5G, WIFI, deviations scope will be more and more less.
Step 1.2, to the data in mobile phone received, utilize special deciphering module, carry out Coordinate Conversion, real time hand after deciphering
The content of machine data includes, (space is by mistake for EPID, TYPE, TIME, LON (true longitude coordinate), LAT (true latitude coordinate), SR
Difference scope).
Step 1.3, based in mobile phone record conversion after LON, LAT value, it is judged that its city, place, give LOC (region) genus
Property, i.e. this mobile phone place City attribution.Traversal data in mobile phone record, LON, LAT after deciphering in being recorded by each of time shaft
Information, uses GIS software (such as ArcMap) to carry out geographical mapping, projects geographical space according to its latitude and longitude coordinates, in conjunction with complete
State city (data smallest particles is prefecture-level city) territory, face figure layer, it is thus achieved that city residing for active user, is entered as this city name
LOC attribute.
Step 1.4, reject at the appointed time section according to LOC in there is not the user of trans-city travel behaviour.Specifically
Rule is: if an EPID at the appointed time in the range of there is two or more LOC, then it represents that this EPID exists trans-city
Travel behaviour, records this EPID, is stored in target database, otherwise abandons this EPID, carries out the differentiation of next user.
In the present embodiment, the tracking data after a certain EPID (e1) deciphering are as shown in the table.In this example, EPID is e1's
User is in different cities on January 1st, 2015 and on March 1st, 2015, therefore e1 is labeled as roamer, lists analysis in
Among object.
Table 1: real-time data in mobile phone newly received after deciphering
Step 2, according to the whole communications records in EPID at the appointed time section T, with LOC for space dividing unit, will
The time that EPID communication activity in each LOC occurs is ranked up, and forms EPID time of communication behavior in each LOC
Axle；On this basis, LON, LAT information in being recorded by each of time shaft, use GIS software (such as ArcMap) to carry out
Reason maps, and projects geographical space according to its latitude and longitude coordinates, generates space-time data；Utilize geographical adjacent_lattice to space-time data
Carry out spatial continuity inspection；Calculate the duration of each time period and the length in this section of path theory in space-time data, with this
Calculate user's gait of march at this section；Obtain user to go on a journey Time-space serial group.
Step 2.1, traversal have been stored in the EPID meeting trans-city trip condition of target database, search in data base
Specify all records of this EPID in time period T.According to the information in every record, with geography in GIS software (such as ArcMap)
Data are mated, and fall in actual map using every record as data point, the information in data point include EPID, TIME,
LON、LAT、LOC。
In this example, the appearance data logger of user e1 is as shown in table 2 behind coupling geographical position.
Table 2: real-time data in mobile phone newly received after deciphering
| RECORDID | EPID | TIME | LON | LAT | LOC |
| …… | …… | …… | …… | …… | …… |
| RECORD(r1-1) | EPID(e1) | 2015-01-02 00:57:00 | 121.303094 | 31.343088 | Shanghai |
| RECORD(r1) | EPID(e1) | 2015-01-02 00:58:00 | 121.295474 | 31.318788 | Shanghai |
| …… | …… | …… | …… | …… | |
| RECORD(r2-1) | EPID(e1) | 2015-03-01 11:30:00 | 119.243514 | 32.353108 | Nanjing |
| RECORD(r2) | EPID(e1) | 2015-03-01 11:32:00 | 119.243884 | 32.356238 | Nanjing |
| …… | …… | …… | …… | …… |
Step 2.2, build the Time-space serial of completely going on a journey of each EPID.For reducing data redundancy, Time-space serial data
Build in units of LOC, only retain the interior Article 1 in each prefecture-level city's continuous communiction record of this EPID at the appointed time section T
Record data point corresponding to FirstRec be the data point that a node is corresponding with the last item record LastRec be a joint
Point, and be ranked up according to the time of origin of every record, build section data.Field in the data of section includes: LNODE
(Time interval spends the time through this section, equal to front for (this section beginning node), NNODE (this section terminal note), TI
The difference of latter two node time), isTRANS (Boolean variable, if trans-regional), DISTANCE (path), SPEED (OK
Enter speed).DISTANCE and the SPEED field of section data can be ignored in this step, and step 2.5 and step are shown in its calculating
2.6.Section and sequence node data are collectively forming full-time null sequence data.
In this example, as a example by user e1, its node data is as shown in table 3.
Table 3:e1 node data
| NODE | TIME | LON | LAT |
| Shanghai | 2015-01-01 00:01:12 | 121.307324 | 31.202432 |
| Shanghai | 2015-03-01 11:45:21 | 121.157453 | 31.398654 |
| Suzhou | 2015-03-01 11:46:32 | 121.121548 | 31.409423 |
| Suzhou | 2015-03-01 12:08:43 | 120.611123 | 31.712098 |
| Wuxi | 2015-03-01 12:09:56 | 120.564536 | 31.712432 |
| Wuxi | 2015-03-01 12:21:20 | 120.165453 | 31.716547 |
| Changzhou | 2015-03-01 12:22:28 | 120.136543 | 31.719306 |
| Changzhou | 2015-03-01 12:32:23 | 119.777423 | 31.891640 |
| Zhenjiang | 2015-03-01 12:33:43 | 119.748534 | 31.919342 |
| Zhenjiang | 2015-03-01 12:41:34 | 119.102543 | 32.035435 |
| Nanjing | 2015-03-01 12:42:11 | 119.056423 | 32.088423 |
| Nanjing | 2015-03-05 12:52:31 | 118.785432 | 32.219543 |
| Beijing | 2015-03-05 15:21:28 | 116.672543 | 40.387123 |
| Beijing | 2015-03-05 18:40:34 | 116.673422 | 40.387875 |
| Harbin | 2015-03-05 22:21:19 | 128.125436 | 45.563453 |
| Harbin | 2015-03-10 15:23:11 | 123.263545 | 45.373124 |
| Shanghai | 2015-03-10 19:00:13 | 121.311454 | 31.167432 |
| Shanghai | 2015-06-30 23:58:28 | 121.354543 | 30.892453 |
Its section data are as shown in table 4.
Table 4:e1 section data
Step 2.3, travel through full-time null sequence data, the labelling departure place-destination's node i.e. gone on a journey therein, below remember
For O-D node, marking convention is:
Statistics EPID is at the distribution of position, incity, each city scatterplot and Annual distribution thereof, if this EPID is in the stop in a certain city
Time, more than Stay_1 (suggestion initial value is 24 hours), the most directly regards this city destination as a stroke, and next
The starting point of Duan Hangcheng, i.e. O-D node；
If EPID in this city time of staying more than Stay_2 (suggestion initial value be 5 hours), but less than Stay_1, and extremely
Be positioned at less non-traffic route, station, airport, Expressway Service measuring point accounting more than 50%, then this city judges it
For O-D point；
If user is at non-traffic route, station, airport, the Stay_R (suggestion initial value is 50 meters) of Expressway Service
Diameter region in continue the time of staying more than Stay_3 (suggestion initial value be 3 hours), then this diameter region is judged to O-D
Point；
In specifying time period T, the head and the tail node location of EPID is accordingly to be regarded as O-D node.
In this example, can labelling Shanghai and Nanjing and Harbin be O-D node.
Step 2.4, the full-time null sequence data of EPID is carried out cutting according to O-D node, formed and a plurality of only have head and the tail joint
Point is the single travel time sequence of O-D node.
In this example, the Time-space serial of e1 is removable to be divided into:
Stroke 1: Shanghai-Nanjing；
Table 5: stroke 1 node data
| NODE | TIME | LON | LAT |
| Shanghai | 2015-03-01 11:45:21 | 121.157453 | 31.398654 |
| Suzhou | 2015-03-01 11:46:32 | 121.121548 | 31.409423 |
| Suzhou | 2015-03-01 12:08:43 | 120.611123 | 31.712098 |
| Wuxi | 2015-03-01 12:09:56 | 120.564536 | 31.712432 |
| Wuxi | 2015-03-01 12:21:20 | 120.165453 | 31.716547 |
| Changzhou | 2015-03-01 12:22:28 | 120.136543 | 31.719306 |
| Changzhou | 2015-03-01 12:32:23 | 119.777423 | 31.891640 |
| Zhenjiang | 2015-03-01 12:33:43 | 119.748534 | 31.919342 |
| Zhenjiang | 2015-03-01 12:41:34 | 119.102543 | 32.035435 |
| Nanjing | 2015-03-01 12:42:11 | 119.056423 | 32.088423 |
Table 6: stroke 1 section data
Stroke 2: Nanjing-Harbin；
Table 7: stroke 2 node data
| NODE | TIME | LON | LAT |
| Nanjing | 2015-03-05 12:52:31 | 118.785432 | 32.219543 |
| Beijing | 2015-03-05 15:21:28 | 116.672543 | 40.387123 |
| Beijing | 2015-03-05 18:40:34 | 116.673422 | 40.387875 |
| Harbin | 2015-03-05 22:21:19 | 128.125436 | 45.563453 |
Table 8: stroke 2 section data
| LNODE | NNODE | TI | IsTRANS |
| Nanjing | Beijing | 2h28s57s | True |
| Beijing | Beijing | 3h19m6s | False |
| Beijing | Harbin | 3h28m35s | True |
| Harbin | Harbin | 4d17h1m58s | False |
Stroke 3: Harbin-Shanghai
Table 9: stroke 3 node data
| NODE | TIME | LON | LAT |
| Harbin | 2015-03-10 15:23:11 | 123.263545 | 45.373124 |
| Shanghai | 2015-03-10 19:00:13 | 121.311454 | 31.167432 |
Table 10: stroke 3 section data
| LNODE | NNODE | TI | IsTRANS |
| Harbin | Shanghai | 3h37m2s | True |
Step 2.5, extracting the air line distance of section data, highway distance and rail distance (need inter-city passenger rail herein
Net and Class III highway network data), i.e. DISTANCE.Concrete grammar can use the nework analysis module in ArcMap, utilizes VBA
Or use ArcEngine to carry out secondary development, batch processing Time-space serial.Obtain three kinds of distances are stored in section sequence data
In.If there is no inter-city railway network and Class III highway network data, also can replace with intercity adjacency matrix.
Intercity adjacency matrix is with city as node, be made up of the shortest path between adjacent cities two-by-two
Transportation network.East China Normal University Wang Zheng teaches seminar in 2008 once according to National Railway Network at that time and Class III highway net
Construct the adjacency matrix between 226 prefecture-level cities of only one China, and calculate the shortest path between Chinese each city with this.
Due to during the calculating of shortest path can on multiple traffic routes any switching laws, therefore in adjacency matrix adjacent cities it
Between distance be generally less than actual highway and rail distance.
Step 2.6, utilize acquired road section length DISTANCE and spend duration TIME, calculate the carrying out in each section
Speed SPEED.When using inter-city railway network and Class III highway network data, it is thus achieved that three kinds of speed: space rate SPEED-S, height
Ferrum speed SPEED-H and network of highways speed SPEED-R；When replacing when using intercity adjacency matrix, adjacency matrix speed
SPEED-M replaces SPEED-H and SPEED-R.
In this example, user e1 is at stroke 1: during Shanghai-Nanjing, four kinds of speed in each section are respectively
Table 11: stroke 1 each section speed
| LNODE | NNODE | SPEED-S | SPEED-H | SPEED-R | SPEED-M |
| Shanghai | Suzhou | 270km/h | 303km/h | 350km/h | 283km/h |
| Suzhou | Suzhou | 235km/h | 290km/h | 342km/h | 245km/h |
| Suzhou | Wuxi | 255km/h | 300km/h | 351km/h | 260km/h |
| Wuxi | Wuxi | 230km/h | 283km/h | 334km/h | 239km/h |
| Wuxi | Changzhou | 280km/h | 305km/h | 368km/h | 285km/h |
| Changzhou | Changzhou | 233km/h | 278km/h | 340km/h | 241km/h |
| Changzhou | Zhenjiang | 248km/h | 295km/h | 358km/h | 252km/h |
| Zhenjiang | Zhenjiang | 239km/h | 280km/h | 340km/h | 243km/h |
| Zhenjiang | Nanjing | 250km/h | 292km/h | 351km/h | 255km/h |
Step 3, according to EPID trip Time-space serial data in SPEED field differentiate non-O-D node node type,
And the trip mode of EPID is finally determined with this.Wherein, the speed per hour of airline carriers of passengers is substantially between 600-750km/h, and high ferro (contains
D prefix motor-car) speed per hour is typically between 200-300km/h, and the F-Zero of highway must not exceed 120km/h.
In step 3.1, first a certain to EPID single trip sequence, the speed SPEED field in each section clusters, choosing
Go out the significant lower section of speed as velocity anomaly section, it is judged that city, place, velocity anomaly section is that intermediate node (includes changing
Take advantage of, stop, refuel, have a rest, go sight-seeing temporarily, traffic congestion etc.), it is judged that standard is as follows:
The first kind: if speed cluster result is 1 class, or different classes of between gaps between their growth rates the least, do not exist significantly
Velocity anomaly point, then this trip sequence is through, abnormal section does not occur；
Equations of The Second Kind: if speed cluster result is more than 2 classes or two classes, and different classes of between speed difference very big, then with
Intersection both sides, friction speed classification section, slow section is abnormal section, and its particular situation can be divided into:
If situation 1 speed cluster result is 2 classes, and abnormity point with discrete formal distribution on Time-space serial (such as table 12
Shown in), then judge that this section is as abnormal section.In table, it is 2 classes that the speed of this sequence can cluster, and C section and H section speed are the lowest
In other sections, it is labeled as abnormal section.
Table 12: cluster result is illustrated
| Section | A | B | C | D | E | F | G | H | I | J | K |
| Speed | 110 | 115 | 20 | 80 | 90 | 85 | 100 | 10 | 110 | 95 | 98 |
If situation 2 cluster result is shown as velocity anomaly point a lot, and two kinds of section is the most substantially continuous
(as shown in table 13), then judging in the boundary section in two class sections, that section slow is abnormal section.In table, C and
D is the boundary of two kinds of speed category, and D section speed is slower, it is determined that for abnormal section.
Table 13: cluster result is illustrated
| Section | A | B | C | D | E | F | G | H | I | J | K |
| Speed | 250 | 280 | 270 | 85 | 80 | 85 | 90 | 100 | 95 | 95 | 98 |
If situation 3 cluster result display speed classification is more than 2, and the section Discrete Distribution that speed is relatively low, and speed is relatively
The substantially continuous distribution of high classification (as shown in table 14), then judge that the slowest that classification of speed is as abnormal section.In table, D road
Duan Sudu is considerably slower than other sections, and between A-C section and E-K section, is then abnormal section.
Table 14: cluster result is illustrated
| Section | A | B | C | D | E | F | G | H | I | J | K |
| Speed | 250 | 280 | 270 | 20 | 80 | 85 | 90 | 100 | 95 | 95 | 98 |
If situation 4 cluster result display speed classification is more than 2, and each speed category the most all presents continuous distribution, then
Judge the slowest that classification of speed as abnormal section, then decision procedure is consistent with situation 2.
If the section of situation 5 friction speed classification is alternately present, there is no obvious rule can follow, then also according to situation 2
Process, i.e. with the slower section of speed, speed separation both sides as abnormal section.
In this example, the speed in stroke 1 Shanghai-Nanjing of user e1 clusters and is shown as 2 classes, but both speed differences
Away from the least, being the most still considered as single stroke, the most significantly transfer or middle stop event do not occur；Stroke 2 Nanjing-Kazakhstan
The speed cluster of your shore is similarly 2 classes, and gaps between their growth rates are the biggest；The speed cluster in stroke 3 Harbin-Shanghai is 1 class.
Step 3.2, determine intermediate node according to velocity anomaly section.Decision rule is:
If head and the tail node LNODE and NNODE in abnormal section is in same city, then this city of labelling is that EPID is in this trip
Intermediate node in activity (include changing to, stop, refuel, have a rest, go sight-seeing temporarily, traffic congestion etc.)；
If head and the tail node LNODE and NNODE in abnormal section be not in same city, then according to this section at LNODE and
The length in NNODE city, takes distance elder as intermediate node.
If how velocity anomaly section across, itself LNODE and NNODE does not adjoins, then cannot differentiate wherein
Which node is intermediate node, abandons mark or abandons this trip Time-space serial.
In this example, user e1 is at stroke 2: during Nanjing-Harbin, four kinds of speed in each section are respectively as follows:
Table 15: stroke 1 each section speed
| LNODE | NNODE | SPEED-S | SPEED-H | SPEED-R | SPEED-M |
| Nanjing | Beijing | 820km/h | 1100km/h | 1210km/h | 1050km/h |
| Beijing | Beijing | 2km/h | 1km/h | 1km/h | 1km/h |
| Beijing | Harbin | 830km/h | 1230km/h | 1265km/h | 1180km/h |
In this example, only there is velocity anomaly section in stroke 2, this section is Beijing-Beijing, therefore, it is determined that during Beijing is
Turn, transfer stop, be intermediate node.
Step 3.3, in addition to O-D node and intermediate node, remaining vertex ticks being on EPID trip route is for passing by
Point.
In this example, the attribute in each city is shown in Table:
Table 16: each City attribution table based on this example
| NODE | TIME |
| Shanghai | O-D point |
| Suzhou | Transit point |
| Wuxi | Transit point |
| Changzhou | Transit point |
| Zhenjiang | Transit point |
| Nanjing | O-D point |
| Beijing | Intermediate node |
| Harbin | O-D point |
Step 3.4, extract the speed in section between O-D node and intermediate node, compare the average speed of each trip mode,
The trip mode (including that civil aviaton, high ferro are containing motor-car, highway) that differentiating each time goes on a journey is used.For avoiding extremum to knot further
The impact of fruit, we take the speed that between O-D node and intermediate node, the median of each section speed is gone on a journey as this.
1, the discrimination standard of civil aviaton is for carrying out speed more than 500km/h, and must there is boat between its stroke head and the tail city
Class (because personal aircraft is few in China, therefore ignore).If gait of march is not navigated more than between 500km/h and head and the tail city
Class, then show that EPID is in other city boarding or landing.Now detection range LNODE and the nearest airport of NNODE carry out having mended,
There is the third party city flight to LNODE or NNODE near if, then between LNODE and NNODE, add airport node, meter
Calculate newly search airport to the distance of LNODE or NNODE nearer apart from it, the travel time of this section is former LNODE to NNODE's
Travel time deducts the theoretic flight time, then judges the vehicles that LNODE or NNODE is used to airport.If nothing
Method searches to obtain suitable flight, then abandon this record.
2, the discrimination standard of high ferro be middle bit rate between 150-500km/h, and the judgement of below 150km/h is public
Go on a journey in road.Wherein, as step 2.5 proposes, if lacking interurban railway net and Class III highway net, intercity adjacent square can be used
Battle array processes, and adjacency matrix is the interurban communication network in a kind of ideal, and the distance that its measuring and calculating obtains is generally less than city
Traffic mileage length actual between city.Additionally, due under practical situation, the tendency of railway is come straight more than highway, if terminal
Identical, its distance walked in practical situations both is less than highway, and the speed of high ferro is far above highway communication, therefore makes
During with intercity adjacency matrix, high ferro and the highway difference in speed can become apparent from the contrary.
In this example, the single stroke speed in Shanghai-Nanjing is at about 300km/h, it is determined that go on a journey for high ferro, Nanjing and north
Capital, Beijing-Harbin and Harbin-Shanghai are spatially discontinuous, and gait of march is more than 500km/h, it is determined that go on a journey for civil aviaton.
Step 4, according to the trip Time-space serial data of a large amount of EPID obtained, and the trip mode supporting with it and joint
Point attribute data, analyzes the flow of intercity migration passenger flow, the flow direction, and the function that each city is in artificial abortion's transition process.
1, the O-D node of all previous trip of all EPID, the flow direction of statistics passenger flow migration and flow are extracted.Thickness with line segment
Represent the size of the volume of the flow of passengers, instruction passenger flow direction, line segment arrow direction.
2, travel amount and trip directional statistics according to the every kind of trip mode in each section migrate passenger flow on detailed path
Flow/flow direction.The same size representing the volume of the flow of passengers with the thickness of line segment, instruction passenger flow direction, line segment arrow direction.
3, adding up in the appearance record of all effective EPID, each city is occurred as O-D point, intermediate node and transit point
The frequency, analyze each city ratio of played the part of different role in passenger flow transition process, analyze each city and migrated in passenger flow
Dominant role in journey.The above results shows with pie ratio chart, with red, blue, black, represent that urban node moves in passenger flow respectively in vain
Play the part of the ratio of starting point, target endpoint, intermediate node, transit point during shifting, represent passenger flow with the size of cake chart diameter
The frequency that in the middle of transition process, this urban node occurs.
In this example, the line direction that goes out of user e1 is from Shanghai-Nanjing-Harbin-Shanghai, for closing stroke, therefore goes up
Sea, Nanjing, Harbin nodal community in, Beijing as intermediate node, Suzhou, Wuxi, Changzhou, Zhenjiang as transit point, meter
Enter above-mentioned city to add up in respective attributes.
In this example, user e1 by high ferro from Shanghai through Suzhou, Wuxi, Changzhou, Zhenjiang to Nanjing, then by civil aviaton from south
Capital to Harbin, finally by civil aviaton from Harbin to Shanghai, counts above-mentioned intercity high ferro or civil aviaton's guest flow statistics through Beijing.
Being drawn population migration flow direction figure by above-mentioned statistics, each city is role attribute figure in population migration is flowed, respectively
Section migrates flowing load diagram.
Claims (7)
1. an intercity migration passenger flow analysing method based on the big data of mobile phone, it is characterised in that comprise the following steps:
Step 1, from common carrier obtain different cities continuous print anonymity on the time and space encryption data in mobile phone, to often
For individual EPID, it is calculated each data in mobile phone in the appointment time period according to the anonymous encryption data in mobile phone of current EPID corresponding
True longitude coordinate LON and true latitude coordinate LAT, different cities is defined as different region LOC, according to calculating
The all of true longitude coordinate LON arrived and true latitude coordinate LAT, it is thus achieved that district residing in current EPID at the appointed time section
The number of territory LOC, the number rejecting region LOC is not more than all EPID of 1, then obtain at the appointed time all across city in section
The EPID in city, and the EPID that these are trans-city is recorded into target database；
Step 2, according to the whole communications records in EPID at the appointed time section each in target database, with region LOC as sky
Between dividing unit, the time that current EPID communication activity in each region LOC occurs is ranked up, forms current EPID
The time shaft of communication behavior in each region LOC；On this basis, the true longitude in being recorded by each of time shaft is sat
Mark LON and true latitude coordinate LAT, carries out geographical mapping, projects geographical space according to its latitude and longitude coordinates, generate space-time number
According to；Utilize geographical adjacent_lattice that space-time data is carried out spatial continuity inspection；Calculate each time period in space-time data time
Long with the length in this section of path theory, calculate user's gait of march at this section with this, it is thus achieved that user goes on a journey Time-space serial group,
Comprise the following steps
Step 2.1, traversal have been stored in the EPID meeting trans-city trip condition of target database, search in target database
All records of current EPID in specifying the time period, the corresponding information in recording every mates with geodata, will be every
Bar record is fallen in actual map as a data point, and each data point at least includes: EPID, record time of origin TIME,
True longitude coordinate LON, true latitude coordinate LAT, region LOC；
Step 2.2, in units of the LOC of region, search and remembered in regional LOC in current EPID at the appointed time section inner segment
The data point corresponding to Article 1 communications records FirstRec of record is as a node and the last item communications records LastRec
Corresponding data point is as a node, and is ranked up, according to the time of origin of every communications records, the node obtained,
Building section data, the field in the data of section at least includes: section beginning node LNODE, section terminal note NNODE, process should
The time TIME that section spends, section data and node data are collectively forming full-time null sequence data；
Step 2.3, travel through full-time null sequence data, labelling departure place therein-destination's node；
Step 2.4, the full-time null sequence data of current EPID is carried out cutting according to departure place-destination's node, formed a plurality of
Only have the single travel time sequence that head and the tail node is departure place-destination's node；
Step 2.5, combine inter-city railway network and Class III highway network data or intercity adjacency matrix extracts the straight of section data
Linear distance, highway distance and rail distance, be path DISTANCE；
Step 2.6, utilize path DISTANCE and time TIME, be calculated speed SPEED that carries out in each section, and
Speed SPEED will be carried out and add in single travel time sequence, when step 2.5 uses inter-city railway network and Class III highway
During network data, it is thus achieved that three kinds of speed: space rate SPEED-S, high ferro speed SPEED-H and network of highways speed SPEED-R；Work as step
When using intercity adjacency matrix in rapid 2.5, it is thus achieved that two kinds of speed: space rate SPEED-S, adjacency matrix speed
SPEED-M；
Step 3, according in the single travel time sequence of current EPID carry out speed SPEED differentiate non-departure place-destination
The node type of node, and the trip mode of current EPID is finally determined with this, comprise the following steps:
Step 3.1, first speed SPEED that carries out in each section in the single travel time sequence of current EPID is clustered,
Select the significant lower section of speed as velocity anomaly section；
Step 3.2, according to the first node LNODE of velocity anomaly section and tail node NNODE, meet one of following condition, then by speed
The abnormal city, place, section of degree is judged as intermediate node:
If the first node LNODE in condition one velocity anomaly section and tail node NNODE are in same city, then this city of labelling is
Intermediate node；
If the first node LNODE in condition two velocity anomaly section and tail node NNODE be not in same city, then according to speed
Abnormal section, in first node LNODE and the length in tail node NNODE city, takes city corresponding to distance elder as middle node
Point；
If how velocity anomaly section across, its first node LNODE and tail node NNODE do not adjoin, then cannot sentence
Wherein which node is intermediate node, abandons mark or abandons current single travel time sequence；
Step 3.3, remaining node in addition to departure place-destination's node and intermediate node in single travel time sequence is all marked
It is designated as transit point；
Step 3.4, extract the speed in section between O-D node and intermediate node, compare the average speed of each trip mode, it determines
The trip mode that each trip is used；
Step 4, according to the single travel time sequence data of a large amount of EPID obtained, and the trip mode supporting with it and joint
Point attribute data, analyzes the flow of intercity migration passenger flow, the flow direction, and the function that each city is in artificial abortion's transition process.
A kind of intercity migration passenger flow analysing method based on the big data of mobile phone, it is characterised in that
Described step 1 comprises the following steps:
Step 1.1, obtaining anonymous data in mobile phone in real time from common carrier, anonymous data in mobile phone includes: EPID, TYPE, TIME,
X, Y, SR, LOC, wherein:
EPID is the unique mobile subscriber identification code in the anonymous One-Way Encryption whole world, be common carrier each user is carried out unidirectional not
Reversible encryption, thus uniquely identify each user, and do not expose Subscriber Number privacy information；
TYPE is current anonymous network action type involved by data in mobile phone；
TIME is the generation moment of current anonymous network action involved by data in mobile phone；
X, Y are to utilize special coordinate encryption method to obtain encryption after the true coordinate position of cellphone subscriber is encrypted calculating to sit
Mark；
SR is spatial dimension, is current anonymous data in mobile phone sterically defined deviations scope；
Step 1.2, to the anonymous data in mobile phone received, utilize special deciphering module to carry out Coordinate Conversion, real-time after deciphering
The content of data in mobile phone includes, EPID, TYPE, TIME, true longitude coordinate LON, true latitude coordinate LAT, space error model
Enclose SR；
Step 1.3, based on the true longitude coordinate LON in real-time data in mobile phone and true latitude coordinate LAT, it is judged that the most real-time
The city, user place that data in mobile phone is corresponding, gives region LOC attribute, travels through the data in mobile phone record of current EPID, when passing through
True longitude coordinate LON and true latitude coordinate LAT in each of countershaft record carry out geographical mapping, according to latitude and longitude coordinates
Project geographical space, in conjunction with territory, face, National urban figure layer, it is thus achieved that city residing for current EPID, this city name is entered as district
Territory LOC attribute；
There is not the EPID of trans-city travel behaviour in step 1.4, specifically advises in rejecting at the appointed time section according to region LOC
Be then: if an EPID at the appointed time in the range of there is two or more regions LOC, then it represents that this EPID exist across city
City's travel behaviour, records this EPID, is stored in target database, otherwise abandons this EPID, carries out the differentiation of next user.
A kind of intercity migration passenger flow analysing method based on the big data of mobile phone, it is characterised in that
In described step 2.3, the flag condition of departure place-destination's node is for meeting one of following condition:
Condition one, add up current EPID position, incity, each city scatterplot distribution and Annual distribution, if this EPID is in a certain city
The time of staying in city is more than time threshold Stay_1, the most directly regards this city destination as a stroke, and next section row
The starting point of journey, is i.e. labeled as departure place-destination's node；
If the current EPID of condition two is in current city time of staying overtime threshold value Stay_2, but less than time threshold Stay_
1, and it is located at least in the node accounting of the described full-time null sequence data of non-traffic route, station, airport, Expressway Service
More than 50%, then current city is labeled as departure place-destination's node；
If condition three user is in non-traffic route, station, airport, the diameter region of length value Stay_R of Expressway Service
Continue time of staying overtime threshold value Stay_3 in territory, then by diameter region, this is judged to departure place-destination's node；
Condition four, current EPID full-time null sequence data in head and the tail node location be accordingly to be regarded as departure place-destination's node.
A kind of intercity migration passenger flow analysing method based on the big data of mobile phone, it is characterised in that
In described step 3.1, the criterion in velocity anomaly section is:
If cluster result is 1 class, or different classes of between gaps between their growth rates the least, there is not obvious velocity anomaly point, then when
Front single travel time sequence is through, abnormal section does not occur；
If cluster result is more than 2 classes or two classes, and different classes of between speed difference very big, then with friction speed classification section
Intersection both sides, slow section is velocity anomaly section.
A kind of intercity migration passenger flow analysing method based on the big data of mobile phone, it is characterised in that
When cluster result is more than 2 classes or two classes, particular situation can be divided into:
If situation 1 cluster result is 2 classes, and the relatively low section of speed with discrete formal distribution in single travel time sequence
On, then the section that speed is relatively low is velocity anomaly section；
If situation 2 cluster result is 2 classes, and the relatively low section of speed is a lot, and the most two kinds of section is the most basic
Continuously, then judge that, in the boundary section in two class sections, that section slow is velocity anomaly section；
If situation 3 cluster result is more than 2 classes, and the section Discrete Distribution that speed is relatively low, and the higher classification of speed connects substantially
Continuous distribution, then judge that the slowest that section of speed is as velocity anomaly section；
If situation 4 cluster result is more than 2 classes, and each speed category the most all presents continuous distribution, then judge that speed is the slowest
That section is velocity anomaly section；
If the section of situation 5 friction speed classification is alternately present, there is no obvious rule can follow, then also according to situation 2 at
Reason, i.e. with the slower section of speed, speed separation both sides for velocity anomaly section.
A kind of intercity migration passenger flow analysing method based on the big data of mobile phone, it is characterised in that
In described step 3.4, take the median of each section speed of departure place-between destination's node and intermediate node as this
The speed of trip:
The discrimination standard of civil aviaton is for carrying out speed more than 500km/h, and must there is flight between its stroke head and the tail city, if row
Enter speed more than 500km/h and head and the tail city between there is no flight, then show current EPID in other city boarding or landing, this
Time detection range head node LNODE and the nearest airport of tail node NNODE carry out having mended, if near there is third party city to first
Node LNODE or the flight of tail node NNODE, then add airport node, meter between first node LNODE and tail node NNODE
Calculate newly search airport is protocercal tail to the first node LNODE nearer apart from it or the distance of tail node NNODE, the travel time of this section
Node LNODE deducts the theoretic flight time to the travel time of tail node NNODE, then judges first node LNODE or tail
The vehicles that node NNODE is used to airport, if suitable flight cannot be searched to obtain, then abandon this record.
The discrimination standard of high ferro be middle bit rate between 150-500km/h, and the judgement of below 150km/h is highway goes out
OK.
A kind of intercity migration passenger flow analysing method based on the big data of mobile phone, it is characterised in that
Described step 4 is analyzed the flow of intercity migration passenger flow, the flow direction, and the functional packet that each city is in artificial abortion's transition process
Include:
Extracting departure place-destination's node of all previous trip of all EPID, the flow direction of statistics passenger flow migration and flow, with line segment
Thickness represents the size of the volume of the flow of passengers, instruction passenger flow direction, line segment arrow direction；
Travel amount according to the every kind of trip mode in each section and trip directional statistics migrate passenger flow flow on detailed path/
Flow to, represent the size of the volume of the flow of passengers, instruction passenger flow direction, line segment arrow direction equally with the thickness of line segment；
Adding up in the appearance record of all effective EPID, each city is as departure place-destination's node, intermediate node and transit point
The frequency occurred, analyzes each city ratio of played the part of different role in passenger flow transition process, analyzes each city in passenger flow
Dominant role in transition process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610367737.6A CN106067154A (en) | 2016-05-30 | 2016-05-30 | A kind of intercity migration passenger flow analysing method based on the big data of mobile phone |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610367737.6A CN106067154A (en) | 2016-05-30 | 2016-05-30 | A kind of intercity migration passenger flow analysing method based on the big data of mobile phone |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106067154A true CN106067154A (en) | 2016-11-02 |
Family
ID=57420851
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610367737.6A Pending CN106067154A (en) | 2016-05-30 | 2016-05-30 | A kind of intercity migration passenger flow analysing method based on the big data of mobile phone |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106067154A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107241512A (en) * | 2017-06-30 | 2017-10-10 | 清华大学 | Intercity Transportation trip mode determination methods and equipment based on data in mobile phone |
| CN107770744A (en) * | 2017-09-18 | 2018-03-06 | 上海世脉信息科技有限公司 | The identification of travelling OD node and hop extracting method under big data environment |
| CN108391265A (en) * | 2018-02-12 | 2018-08-10 | 中国联合网络通信集团有限公司 | A kind of determining method and device for roaming the user that passes by |
| CN109889988A (en) * | 2017-12-06 | 2019-06-14 | 北京亿阳信通科技有限公司 | Method and apparatus based on subway scene communications records analysis Network status |
| CN110046174A (en) * | 2019-03-07 | 2019-07-23 | 特斯联（北京）科技有限公司 | A kind of population migration analysis method and system based on big data |
| CN110728433A (en) * | 2019-09-19 | 2020-01-24 | 重庆市交通规划研究院 | Land parcel resident population measuring and calculating method based on mobile phone signaling |
| CN111770452A (en) * | 2020-05-27 | 2020-10-13 | 中山大学 | Mobile phone signaling stop point identification method based on personal travel track characteristics |
| CN112383878A (en) * | 2020-09-27 | 2021-02-19 | 中国信息通信研究院 | Collaborative computing method and electronic device |
| CN112434101A (en) * | 2020-11-23 | 2021-03-02 | 北京航空航天大学 | System for carrying out people flow migration analysis by using shared trip big data |
| CN112689301A (en) * | 2019-10-17 | 2021-04-20 | 中国移动通信集团陕西有限公司 | Switching sequence identification method and device of road scene and computing equipment |
| CN112711576A (en) * | 2020-12-11 | 2021-04-27 | 上海城市交通设计院有限公司 | Method for identifying inter-city travel modes of high-speed rail and airplane with mobile phone signaling data |
| 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 |
| CN113495911A (en) * | 2020-03-18 | 2021-10-12 | 百度在线网络技术（北京）有限公司 | Migration information processing 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 |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102595323A (en) * | 2012-03-20 | 2012-07-18 | 北京交通发展研究中心 | Method for obtaining resident travel characteristic parameter based on mobile phone positioning data |
| CN105389996A (en) * | 2015-12-17 | 2016-03-09 | 北京亚信蓝涛科技有限公司 | Traffic operation condition characteristic parameter extraction method based on big data |
| CN105488120A (en) * | 2015-11-23 | 2016-04-13 | 上海川昱信息科技有限公司 | Method for collecting spatial population distribution in real time on basis of mobile phone big data and realizing large passenger flow early warning |
-
2016
- 2016-05-30 CN CN201610367737.6A patent/CN106067154A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102595323A (en) * | 2012-03-20 | 2012-07-18 | 北京交通发展研究中心 | Method for obtaining resident travel characteristic parameter based on mobile phone positioning data |
| CN105488120A (en) * | 2015-11-23 | 2016-04-13 | 上海川昱信息科技有限公司 | Method for collecting spatial population distribution in real time on basis of mobile phone big data and realizing large passenger flow early warning |
| CN105389996A (en) * | 2015-12-17 | 2016-03-09 | 北京亚信蓝涛科技有限公司 | Traffic operation condition characteristic parameter extraction method based on big data |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10674315B2 (en) | 2017-06-30 | 2020-06-02 | Shandong Provincial Communications Planning And Design Institute | Method and device for judging intercity transportation mode based on mobile phone data |
| WO2019001045A1 (en) * | 2017-06-30 | 2019-01-03 | 山东省交通规划设计院 | Intercity traffic mode determining method and device based on mobile phone data |
| CN107241512A (en) * | 2017-06-30 | 2017-10-10 | 清华大学 | Intercity Transportation trip mode determination methods and equipment based on data in mobile phone |
| CN107241512B (en) * | 2017-06-30 | 2019-10-18 | 清华大学 | Intercity Transportation trip mode judgment method and equipment based on data in mobile phone |
| CN107770744A (en) * | 2017-09-18 | 2018-03-06 | 上海世脉信息科技有限公司 | The identification of travelling OD node and hop extracting method under big data environment |
| CN109889988A (en) * | 2017-12-06 | 2019-06-14 | 北京亿阳信通科技有限公司 | Method and apparatus based on subway scene communications records analysis Network status |
| CN108391265A (en) * | 2018-02-12 | 2018-08-10 | 中国联合网络通信集团有限公司 | A kind of determining method and device for roaming the user that passes by |
| CN108391265B (en) * | 2018-02-12 | 2020-12-22 | 中国联合网络通信集团有限公司 | Method and device for determining roaming transit user |
| CN110046174A (en) * | 2019-03-07 | 2019-07-23 | 特斯联（北京）科技有限公司 | A kind of population migration analysis method and system based on big data |
| CN110728433A (en) * | 2019-09-19 | 2020-01-24 | 重庆市交通规划研究院 | Land parcel resident population measuring and calculating method based on mobile phone signaling |
| CN110728433B (en) * | 2019-09-19 | 2023-05-26 | 重庆市交通规划研究院 | Land occupation population measuring and calculating method based on mobile phone signaling |
| CN112689301A (en) * | 2019-10-17 | 2021-04-20 | 中国移动通信集团陕西有限公司 | Switching sequence identification method and device of road scene and computing equipment |
| CN112689301B (en) * | 2019-10-17 | 2024-04-09 | 中国移动通信集团陕西有限公司 | Road scene switching sequence identification method and device and computing equipment |
| CN113495911A (en) * | 2020-03-18 | 2021-10-12 | 百度在线网络技术（北京）有限公司 | Migration information processing method and device, electronic equipment and storage medium |
| CN111770452A (en) * | 2020-05-27 | 2020-10-13 | 中山大学 | Mobile phone signaling stop point identification method based on personal travel track characteristics |
| CN112383878A (en) * | 2020-09-27 | 2021-02-19 | 中国信息通信研究院 | Collaborative computing method and electronic device |
| CN112434101A (en) * | 2020-11-23 | 2021-03-02 | 北京航空航天大学 | System for carrying out people flow migration analysis by using shared trip big data |
| CN112434101B (en) * | 2020-11-23 | 2021-06-25 | 北京航空航天大学 | System for carrying out people flow migration analysis by using shared trip big data |
| CN112711576B (en) * | 2020-12-11 | 2023-03-10 | 上海城市交通设计院有限公司 | Method for identifying inter-city travel modes of high-speed rail and airplane with mobile phone signaling data |
| CN112711576A (en) * | 2020-12-11 | 2021-04-27 | 上海城市交通设计院有限公司 | Method for identifying inter-city travel modes of high-speed rail and airplane with mobile phone signaling data |
| 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 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106067154A (en) | A kind of intercity migration passenger flow analysing method based on the big data of mobile phone | |
| CN108564226B (en) | Bus route optimization method based on taxi GPS and mobile phone signaling data | |
| Yang et al. | Scalable space-time trajectory cube for path-finding: A study using big taxi trajectory data | |
| CN110298500A (en) | A kind of urban transportation track data set creation method based on taxi car data and city road network | |
| CN106096631A (en) | A kind of recurrent population's Classification and Identification based on the big data of mobile phone analyze method | |
| CN105809962A (en) | Traffic trip mode splitting method based on mobile phone data | |
| CN110555992B (en) | Taxi driving path information extraction method based on GPS track data | |
| CN110555544B (en) | Traffic demand estimation method based on GPS navigation data | |
| CN105117595A (en) | Floating car data based private car travel data integration method | |
| CN105844031B (en) | A kind of urban transportation gallery recognition methods based on mobile phone location data | |
| Kreimeier et al. | Market volume estimation of thin-haul On-Demand Air Mobility services in Germany | |
| Yuan et al. | Recognition of functional areas based on call detail records and point of interest data | |
| Ge et al. | Urban taxi ridership analysis in the emerging metropolis: Case study in Shanghai | |
| CN115100012A (en) | Method for calculating walking accessibility of rail transit station | |
| CN110413855A (en) | A kind of region entrance Dynamic Extraction method based on taxi drop-off point | |
| Kashiyama et al. | Pseudo-PFLOW: Development of nationwide synthetic open dataset for people movement based on limited travel survey and open statistical data | |
| Hatziioannidu et al. | Passenger Demand And Patterns Of Tourists’ Mobility In The Aegean Archipelago With Combined Use Of Big Datasets From Mobile Phones And Statistical Data From Ports And Airports | |
| Hu et al. | Dynamic modeling of urban population travel behavior based on data fusion of mobile phone positioning data and FCD | |
| Jiang et al. | Air Corridor Evaluation with Purpose-specific Metrics | |
| Wang et al. | Understanding the Shortest Route Selection Behavior for Private Cars Using Trajectory Data and Navigation Information | |
| CN114666738A (en) | Territorial space planning method and system based on mobile phone signaling | |
| Lei et al. | Research on multi-objective bus route planning model based on taxi GPS data | |
| Jiang et al. | A Metrics-based Method for Evaluating Corridors for Urban Air Mobility Operations | |
| Sabet et al. | A cost-effective methodology to compare travel time and speed: a tale of 11 cities | |
| CN106485921A (en) | A kind of region highway network wagon flow state analysis method based on social data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20161102 |
|
| WD01 | Invention patent application deemed withdrawn after publication |