CN108427965A - A kind of hot spot region method for digging based on road network cluster - Google Patents
A kind of hot spot region method for digging based on road network cluster Download PDFInfo
- Publication number
- CN108427965A CN108427965A CN201810179464.1A CN201810179464A CN108427965A CN 108427965 A CN108427965 A CN 108427965A CN 201810179464 A CN201810179464 A CN 201810179464A CN 108427965 A CN108427965 A CN 108427965A
- Authority
- CN
- China
- Prior art keywords
- cluster
- point
- road network
- data
- hot spot
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/9038—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Traffic Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A kind of trip hot spot region method for digging based on road network trajectory clustering is claimed in the present invention.In the method, by taxi trajectory map to road network, and the clustering method combined using collected point of interest in real road and track.In conjunction with density peaks clustering algorithm, it is proposed that optimize the OPAM algorithms of initial center, i.e. DP OPAM based on density peaks.Algorithm, to the shortest distance of higher density points, picks out density higher and the classification belonging to the nearest data point, as initial cluster center using the local density of data point and these points using decision diagram.Cluster result is obtained using the OPAM clustering algorithms for increasing backward learning according to initial cluster center.New algorithm and original OPAM algorithms are compared, new algorithm can not only automatically determine cluster centre, and improve accuracy rate and cluster time, realize user's trip hot spot region analysis.
Description
Technical field
The invention belongs to a kind of data digging method more particularly to a kind of taxi trajectory clustering sides based on road network
Method.
Background technology
The hot spot that intelligent transportation develops as world today's communications and transportation, while supporting traffic management, more
Emphasis meets the needs of common people's trip and public's traffic trip.In recent years, Intelligent Transport Systems Construction rapidly developed, many elder generations
Into technology be widely used in intelligent transportation system.The extensive use of GPS device makes the extraction of track become more convenient.This
A little GPS devices can be collected into a large amount of shift position sequence information and on-board status information, these data contain abundant
Traffic information and user behavior information.By being analyzed track data and being excavated, we can understand traffic, rationally
It plans stroke, finds crowd's behavioural characteristic, assist to improve traffic etc..
Urban road network traffic can be covered all around by hiring out wheel paths, can reflect real-time high traffic degree and circulation
Degree, can also reflect the trip rule and provincial characteristics of crowd.So by dividing the mass data for hiring out wheel paths
Analysis finds the profound information being hidden in data, by means of data mining technology, analyzes the description of data global feature and hands over
On-state gesture development prediction carries out Vehicle Detection for vehicle supervision department and road control provides support etc. and plays and focuses on your writing
With.
Clustering as a kind of common data mining technology, can as the tool for the distribution situation for obtaining data,
Convenient for observing the feature of each cluster data, concentrates and the collection cooperation that specifically clusters further is analyzed.Further, it is also possible to as it
The pre-treatment step of his algorithm (such as classification and qualitative inductive algorithm).The trajectory clustering of mobile object, by finding similar fortune
The modes such as dynamic rail mark, extraction motion feature, find the characteristics of motion and behavior pattern of mobile object.The track of taxi be by
Disconnected sequence of points is constituted.The traditional clustering in track when measuring track similitude, consider mostly when between points
Air line distance, and the distance of reality is had ignored up to situation.
The clustering research of track of vehicle, there are mainly two types of methods：One is divided whole track as object
Class compares, another then be that track is divided into sub-trajectory section according to certain standard, classifies to obtained sub-trajectory section.Before
The advantages of person, is that method is simple, and convenient for the similitude between intuitive evaluation track, but simultaneously, this method cannot be good
The local feature of track is told, Clustering Effect is usually not ideal enough.It is special in track part can to improve the former for later approach
The problem of being brought in terms of sign, for track of different shapes, Clustering Effect is more preferably.But the disadvantage is that the method for track segmentation is to poly-
Class result is affected, and different dividing methods may cause the widely different of result.
Invention content
Present invention seek to address that the above problem of the prior art.It proposes one kind and being remarkably improved Clustering Effect, realize and use
The hot spot region method for digging based on road network cluster that family trip region is excavated.Technical scheme is as follows：
A kind of hot spot region method for digging based on road network cluster comprising following steps：
Step 1：Taxi track data collection is collected, carries out including data normalization, normalized data prediction, retain
Effective field deletes redundant data, obtains pretreated vehicle on-board and off-board tracing point；
Step 2：Determine city longitude and latitude range, that the city is extracted on website is emerging including market, school increasing income
Interesting point；
Step 3：The road network information for obtaining city, tracing point is mapped in road network；
Step 4：It chooses and is used as training set by 80% in the pretreated vehicle on-board and off-board tracing point of step 1, use
The improved OPAM algorithms for being optimized initial center based on density peaks are clustered out representative and got on or off the bus the region of hot spot, and improvement is main
It is：Using density peak choose initial cluster center, the selection of initial point is more acurrate, convenient, remaining 20% be used as test set, survey
The Clustering Effect of model is put up in examination by 80% in on-board and off-board tracing point as training set；
Step 5：By the collected point of interest with road network information of input step 2 in the model of step 4, cluster is had
Cluster result and collected point of interest are compared, judge the heat of resident trip by the resident's thermal point structure region for having road network feature
Point region.
Further, the step 1 is specially：The taxi track data collection for collecting city month first, chooses the city
City's data volume more concentrates one week track data, carries out data prediction, retains tracing point longitude and latitude degrees of data of getting on or off the bus, up and down
The effective fields such as vehicle time data delete redundant data.
Further, the step 2 determines city longitude and latitude range, increase income extracted on website the city include market,
Point of interest including school, specially：
First in the longitude and latitude range for inputting target cities on the openstreetmap of website of increasing income, entire city is downloaded
Map, way represents the motion track of user, node delegated paths in derived OSM map datums.Choosing node labels is
Residence, school, shop are to represent point of interest.
Further, the step 3 obtains the road network information in city, tracing point is mapped in road network, specially：
Electronic map data is obtained using TAREEG network service items, extracts the road network information in the city, extracts city
After road net data, as ST-Matching models will be above-mentioned obtained by GPS motion tracks project the road network map got
On, driver is obtained by (j-1+1) a continuous moment p on the e of each sectioni,…,pjTracing point.
Further, the step 4 is specially：80% conduct in the vehicle on-board and off-board tracing point handled well is chosen first
Training set clusters out representative around central point partition clustering algorithm (OPAM) based on backward learning and gets on or off the bus hot spot using improving
Region improves OPAM algorithms and is divided into three phases：First stage initializes, and constructs decision diagram, chooses far from most of sample
Upper right comer region density peaks point as initial cluster center, density peaks point number is class number of clusters k；Second stage constructs
Initial cluster center calculates the minimum range of each point and each cluster centre in data set, remaining sample point is assigned to most
Nearly initial classes cluster center, forms initial division, calculates cluster error sum of squares；Phase III backward learning simultaneously substitutes into the center of surrounding
Point divides clustering algorithm (PAM), by k cluster that typical PAM clustering algorithms obtain and obtain after backward learning k reversely clusters into
Row permutation and combination obtains k × k class cluster combination, finds the maximum class cluster combination of silhouette coefficient.
Further, the step of PAM algorithms are as follows：
(1) arbitrarily choose k element from data-oriented collection D, by k selected rubidium marking be initially represent object or
Seed oj；
(2) according to Euclidean distance calculation, calculate in data set D any non-represents object oiObject is represented with k
The distance between, and by oiIt is assigned to and its cluster representated by the nearest representative object；
(3) arbitrary selection one is non-represents object orandom；
(4) total cost S is calculated：
S=dist (p, orandom)-dist(p,oj)
(5) if total cost S ＜ 0, show non-to represent object orandomIt is more excellent solution, element orandomIt can replace element
oj, the new k set for representing object are formed, step (2) is continued back to, do the object distribution of a new round；
(6) if total cost S ＞ 0, show to represent object ojIt is more excellent solution, goes to step (3), chooses non-representative pair again
As carrying out the comparison of total cost, until sending cost S no longer to change to get to k class cluster of total cost minimum.
It advantages of the present invention and has the beneficial effect that：
The present invention by the trip analysis of central issue combination road network of resident, using in specific road network interest region and
The method that original clustering cluster is combined cluster, original cluster gather the interest provincial characteristics for including in new cluster and represent resident again
The hot spot region of trip solves in theorem in Euclid space insufficient existing for time, space aspect.This method, which uses, is based on density peaks
The method construct decision tree of optimization initial center determines initial center, reduces calculation amount and makes cluster accuracy rate higher.And
Cluster is combined by particular interest point and track, solves the problems, such as that Deta sparseness and calculation amount are huge, realizes the track of user
Behavioural analysis.
Description of the drawings
Fig. 1 is that the present invention provides preferred embodiment PAM clustering algorithm flow charts；
Fig. 2 OPAM clustering algorithm flow charts.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, detailed
Carefully describe.Described embodiment is only a part of the embodiment of the present invention.
The present invention solve above-mentioned technical problem technical solution be：
As shown in Fig. 2, the invention uses the OPAM clustering algorithms and road network for optimizing initial center based on density peaks
In conjunction with comprising the concrete steps that for progress hot spot region method for digging：
Step 1：The taxi track data collection for collecting city month, chooses the Urban Data amount and more concentrates one week
Track data.Data prediction is carried out, reservation is got on or off the bus tracing point longitude and latitude degrees of data, the effective fields such as time data of getting on or off the bus,
Delete redundant data.
Step 2：In the longitude and latitude range for inputting target cities on the openstreetmap of website of increasing income, entire city is downloaded
Map.Way represents the motion track of user, node delegated paths in derived OSM map datums.Due in OSM source datas
Way object records are user's motion tracks, so way is not only to indicate road information, can also indicate architecture information, such as：
The exterior contour in one building.So we are also required to carry out corresponding screening and filtering to way objects, by unwanted way information
Remove.Still further aspect, the node sampled points in the map source data of download are very more, only needed in actual path planning process
It is to be understood that the crossroad place intersected between way and way.Having in way and node much represents different attribute
In tag, way a reserved property be highway attribute values be residential, service, living_street,
unclassified、trunk、trunk_link、secondary、secondary_link、primary、tertiary、
The different way but identical nodes of node are deleted in tertiary_link, node.Selection node labels be residence,
School, shop are to represent point of interest.
Step 3：
Which section is the GPS track of the driver of above-mentioned acquisition conceal driver's each moment in comprising momentary position coordinates
On information.Electronic map data is obtained using TAREEG network service items, extracts the road network information in the city.Extract city
After road net data, as ST-Matching models will be above-mentioned obtained by GPS motion tracks project the road network map got
On, driver is obtained by (j-1+1) a continuous moment p on the e of each sectioni,…,pjTracing point.
Step 4
80% chosen in the vehicle on-board and off-board tracing point handled well is used as training set, is clustered out using OPAM algorithms are improved
Represent the region for hot spot of getting on or off the bus.DP-OPAM algorithms are divided into three phases：Initialization, construction initial cluster center and reversed
It practises and substitutes into PAM algorithms.
If the data point of sample is i, local density ρi, the local density ρ of data point iiCalculation be：Wherein,dcTo block distance, for mass data, local density is real
Matter is the relative density between data point, so dcWith robustness.Define δiIt is data point i to any point bigger than its density
Distance minimum value：δi=minj:ρ j ＞ ρ i(dij) for the maximum point of local density, specially treated is needed, the value generally changed the time
For：δi=maxj(dij)
First stage initializes
(1) initialization finds out the distance between each data point matrix D={ dijI, j=1 ..., n, and determine block away from
From.
(2) according to formula S=dist (p, orandom)-dist(p,oj) local density is found out, utilize formulaCalculate the high density distance δ of samplei。
(3) construction ρ selects local density ρ and the higher numbers of high density distance δ for horizontal axis, to be the decision diagram of the δ longitudinal axis
Strong point, and be significantly away from most of sample upper right comer region density peaks point as initial cluster center, density peaks point
Number is class number of clusters k.
Second stage constructs initial cluster center
(1) minimum range for calculating each point and each cluster centre in data set, remaining sample point is assigned to recently
Initial classes cluster center forms initial division, calculates cluster error sum of squares.
Phase III backward learning simultaneously substitutes into PAM algorithms
(1) k original cluster obtained above is subjected to backward learning, acquires k corresponding reversed clusters.
(2) it by k cluster that typical PAM clustering algorithms obtain and k reversed clusters is obtained after backward learning carries out arrangement group
Conjunction obtains k × k class cluster combination.
(3) spacing b (o) and silhouette coefficient s (o) between spacing a (o), cluster are calculated in the cluster of each cluster class combination, is compared
s1,s2,...,sk×k, find the maximum class cluster combination of silhouette coefficient.
Wherein, the step of PAM algorithms are as follows：
(1) arbitrarily choose k element from data-oriented collection D, by k selected rubidium marking be initially represent object or
Seed oj；
(2) according to Euclidean distance calculation, calculate in data set D any non-represents object oiObject is represented with k
The distance between, and by oiIt is assigned to and its cluster representated by the nearest representative object；
(3) arbitrary selection one is non-represents object orandom；
(4) total cost S is calculated：
S=dist (p, orandom)-dist(p,oj)
(5) if total cost S ＜ 0, show non-to represent object orandomIt is more excellent solution, element orandomIt can replace element
oj, the new k set for representing object are formed, step (2) is continued back to, do the object distribution of a new round；
(6) if total cost S ＞ 0, show to represent object ojIt is more excellent solution, goes to step (3), chooses non-representative pair again
As carrying out the comparison of total cost, until sending cost S no longer to change to get to k class cluster of total cost minimum.
Step 5
By the collected point of interest with road network information of input step 2 in the model of step 4, the k that cluster result obtains
A class cluster and collected representative point of interest carry out similarity measurement, which interest analysis resident trip region belongs to
Point, to analyze resident hot spot region.Cluster obtains resident's thermal point structure region with road network feature.By cluster result and adopt
The point of interest comparison collected, judges the hot spot region of resident trip.
For track tr to be measuredaAnd trbUsing Hausdorff range measurements track similarity.H(tra,trb)=max
{h(tra,trb),h(trb,tra), wherein It calculates in two tracks and is each put on an other track using Hausdorff distances
Then the minimum value of all the points is found out maximum from respective minimum value set.Think when less than similarity threshold with it is emerging
It is similar in the interesting space of points, it will be saved in candidate collection.The track for distance in candidate collection being more than some threshold value is deleted, and is obtained
The road network point of interest nearest to the mark that leaves the right or normal track, i.e. resident trip hot spot region.
The above embodiment is interpreted as being merely to illustrate the present invention rather than limit the scope of the invention.
After the content for having read the record of the present invention, technical staff can make various changes or modifications the present invention, these equivalent changes
Change and modification equally falls into the scope of the claims in the present invention.
Claims (6)
1. a kind of hot spot region method for digging based on road network cluster, which is characterized in that include the following steps：
Step 1：Taxi track data collection is collected, carries out including data normalization, normalized data prediction, retain effective
Field deletes redundant data, obtains pretreated vehicle on-board and off-board tracing point；
Step 2：City longitude and latitude range is determined, in the point of interest for extracting the city on website including market, school of increasing income；
Step 3：The road network information for obtaining city, tracing point is mapped in road network；
Step 4：It chooses and is used as training set by 80% in the pretreated vehicle on-board and off-board tracing point of step 1, using improvement
The OPAM algorithms for optimizing initial center based on density peaks cluster out representative and get on or off the bus the region of hot spot, improvement mainly exists
In：Initial cluster center is chosen using density peak；Remaining 20% be used as test set, test by on-board and off-board tracing point 80% make
The Clustering Effect of model is put up for training set；
Step 5：By the collected point of interest with road network information of input step 2 in the model of step 4, cluster is obtained with road
Cluster result and collected point of interest are compared, judge the hot zone of resident trip by resident's thermal point structure region of net feature
Domain.
2. the hot spot region method for digging according to claim 1 based on road network cluster, which is characterized in that the step 1
Specially：The taxi track data collection for collecting city month first, chooses the track that the Urban Data amount is more concentrated one week
Data carry out data prediction, retain tracing point longitude and latitude degrees of data of getting on or off the bus, and the effective fields such as time data of getting on or off the bus are deleted
Redundant data.
3. the hot spot region method for digging according to claim 1 based on road network cluster, which is characterized in that the step 2
Determine city longitude and latitude range, in the point of interest for extracting the city on website including market, school of increasing income, specially：
First in the longitude and latitude range for inputting target cities on the openstreetmap of website of increasing income, the map in entire city is downloaded,
Way represents the motion track of user, node delegated paths in derived OSM map datums.Choosing node labels is
Residence, school, shop are to represent point of interest.
4. the hot spot region method for digging according to claim 1 based on road network cluster, which is characterized in that the step 3
The road network information for obtaining city, tracing point is mapped in road network, specially：
Electronic map data is obtained using TAREEG network service items, extracts the road network information in the city, extracts city road network
After data, as ST-Matching models will be above-mentioned obtained by GPS motion tracks project in the road network map got, obtain
To driver by (j-1+1) a continuous moment p on the e of each sectioni,…,pjTracing point.
5. the hot spot region method for digging according to claim 1 based on road network cluster, which is characterized in that
The step 4 is specially：80% chosen first in the vehicle on-board and off-board tracing point handled well is used as training set, using changing
It gets on or off the bus the region of hot spot into representative is clustered out around central point partition clustering algorithm OPAM based on backward learning, improves OPAM and calculate
Method is divided into three phases：First stage initializes, and constructs decision diagram, chooses the close of the upper right comer region far from most of sample
Peak point is spent as initial cluster center, and density peaks point number is class number of clusters k；Second stage constructs initial cluster center, meter
Remaining sample point is assigned to nearest initial classes cluster center by the minimum range to count according to each point and each cluster centre of concentration,
Initial division is formed, cluster error sum of squares is calculated；Phase III backward learning is simultaneously substituted into around central point partition clustering algorithm
PAM by k cluster that typical PAM clustering algorithms obtain and after backward learning obtains k reversed clusters and carries out permutation and combination obtaining k
The maximum class cluster combination of silhouette coefficient is found in × k class cluster combination.
6. the hot spot region method for digging according to claim 5 based on road network cluster, which is characterized in that
The step of PAM algorithms, is as follows：
(1) k element is arbitrarily chosen from data-oriented collection D, is initially to represent object or seed by k selected rubidium marking
oj；
(2) according to Euclidean distance calculation, calculate in data set D any non-represents object oiIt is represented between object with k
Distance, and by oiIt is assigned to and its cluster representated by the nearest representative object；
(3) arbitrary selection one is non-represents object orandom；
(4) total cost S is calculated：
S=dist (p, orandom)-dist(p,oj)
(5) if total cost S ＜ 0, show non-to represent object orandomIt is more excellent solution, element orandomIt can replace element oj, formed
The k new set for representing object, continue back to step (2), do the object distribution of a new round；
(6) if total cost S ＞ 0, show to represent object ojBe more excellent solution, go to step (3), choose again it is non-represent object into
The comparison of row total cost, until sending cost S no longer to change to get to k class cluster of total cost minimum.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810179464.1A CN108427965B (en) | 2018-03-05 | 2018-03-05 | Hot spot area mining method based on road network clustering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810179464.1A CN108427965B (en) | 2018-03-05 | 2018-03-05 | Hot spot area mining method based on road network clustering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108427965A true CN108427965A (en) | 2018-08-21 |
| CN108427965B CN108427965B (en) | 2022-08-23 |
Family
ID=63157793
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810179464.1A Active CN108427965B (en) | 2018-03-05 | 2018-03-05 | Hot spot area mining method based on road network clustering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108427965B (en) |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109408562A (en) * | 2018-11-07 | 2019-03-01 | 广东工业大学 | A kind of grouping recommended method and its device based on client characteristics |
| CN109739585A (en) * | 2018-12-29 | 2019-05-10 | 广西交通科学研究院有限公司 | The traffic congestion point discovery method calculated based on spark cluster parallelization |
| CN110399686A (en) * | 2019-07-30 | 2019-11-01 | 中国民航大学 | A kind of unrelated aircraft flight profiles clustering method of parameter based on silhouette coefficient |
| CN110427444A (en) * | 2019-07-26 | 2019-11-08 | 北京百度网讯科技有限公司 | Navigation guide point method for digging, device, equipment and storage medium |
| CN110609824A (en) * | 2019-09-09 | 2019-12-24 | 南京师范大学 | Hot spot area detection method based on dynamic space network model under urban road network environment |
| CN110726418A (en) * | 2019-10-10 | 2020-01-24 | 北京百度网讯科技有限公司 | Method, device and equipment for determining interest point region and storage medium |
| CN110849379A (en) * | 2019-10-23 | 2020-02-28 | 南通大学 | Access and exit traffic state symbol expression method for navigation map |
| CN110866559A (en) * | 2019-11-14 | 2020-03-06 | 上海中信信息发展股份有限公司 | Poultry behavior analysis method and device |
| CN111121803A (en) * | 2019-11-27 | 2020-05-08 | 北京中交兴路信息科技有限公司 | Method and device for acquiring common stop points of road |
| CN111275962A (en) * | 2019-12-30 | 2020-06-12 | 深圳市麦谷科技有限公司 | Vehicle track data aggregation effect prediction method and device |
| CN111708853A (en) * | 2020-05-25 | 2020-09-25 | 安徽师范大学 | Taxi hot spot region extraction method based on characteristic density peak clustering |
| WO2020191876A1 (en) * | 2019-03-26 | 2020-10-01 | 中国电子科技集团公司第二十八研究所 | Hotspot path analysis method based on density clustering |
| CN112070179A (en) * | 2020-09-22 | 2020-12-11 | 中国人民解放军国防科技大学 | Adaptive space-time trajectory clustering method based on density peak value |
| CN112163370A (en) * | 2020-09-18 | 2021-01-01 | 国网冀北电力有限公司承德供电公司 | Power distribution network line loss calculation method based on improved ASMDE algorithm |
| CN113158817A (en) * | 2021-03-29 | 2021-07-23 | 南京信息工程大学 | Objective weather typing method based on rapid density peak clustering |
| CN113611115A (en) * | 2021-08-06 | 2021-11-05 | 安徽师范大学 | Vehicle track clustering method based on road network sensitive characteristics |
| CN113869465A (en) * | 2021-12-06 | 2021-12-31 | 深圳大学 | I-nice algorithm optimization method, device, equipment and computer readable storage medium |
| CN114120018A (en) * | 2020-08-25 | 2022-03-01 | 四川大学 | Space activity quantification method based on crowd clustering track entropy |
| CN114485696A (en) * | 2021-12-23 | 2022-05-13 | 高德软件有限公司 | Method for acquiring arrival point, electronic device and storage medium |
| CN114677048A (en) * | 2022-04-22 | 2022-06-28 | 北京阿帕科蓝科技有限公司 | Demand area mining method |
| CN114999151A (en) * | 2022-05-24 | 2022-09-02 | 电子科技大学 | Urban traffic flow hierarchical analysis method and device based on density in GPS track |
| CN112070179B (en) * | 2020-09-22 | 2024-05-31 | 中国人民解放军国防科技大学 | Self-adaptive space-time track clustering method based on density peak value |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104167092A (en) * | 2014-07-30 | 2014-11-26 | 北京市交通信息中心 | Method and device for determining taxi pick-up and drop-off hot spot region center |
| CN105825672A (en) * | 2016-04-11 | 2016-08-03 | 中山大学 | City guidance area extraction method based on floating car data |
| CN107301254A (en) * | 2017-08-24 | 2017-10-27 | 电子科技大学 | A kind of road network hot spot region method for digging |
-
2018
- 2018-03-05 CN CN201810179464.1A patent/CN108427965B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104167092A (en) * | 2014-07-30 | 2014-11-26 | 北京市交通信息中心 | Method and device for determining taxi pick-up and drop-off hot spot region center |
| CN105825672A (en) * | 2016-04-11 | 2016-08-03 | 中山大学 | City guidance area extraction method based on floating car data |
| CN107301254A (en) * | 2017-08-24 | 2017-10-27 | 电子科技大学 | A kind of road network hot spot region method for digging |
Non-Patent Citations (4)
| Title |
|---|
| GUOQING QIU ET AL: "An improved clustering algorithm based on reverse learning in intelligent transportation", 《AIP CONFERENCE PROCEEDINGS 1839》 * |
| 冯琦森: "基于出租车轨迹的居民出行热点路径和区域挖掘", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 程静等: "基于时间序列聚类方法分析北京出租车出行量的时空特征", 《地球信息科学学报》 * |
| 谢娟英等: "密度峰值优化初始中心的K-medoids聚类算法", 《计算机科学与探索》 * |
Cited By (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109408562A (en) * | 2018-11-07 | 2019-03-01 | 广东工业大学 | A kind of grouping recommended method and its device based on client characteristics |
| CN109739585A (en) * | 2018-12-29 | 2019-05-10 | 广西交通科学研究院有限公司 | The traffic congestion point discovery method calculated based on spark cluster parallelization |
| CN109739585B (en) * | 2018-12-29 | 2022-02-18 | 广西交通科学研究院有限公司 | Spark cluster parallelization calculation-based traffic congestion point discovery method |
| WO2020191876A1 (en) * | 2019-03-26 | 2020-10-01 | 中国电子科技集团公司第二十八研究所 | Hotspot path analysis method based on density clustering |
| CN110427444A (en) * | 2019-07-26 | 2019-11-08 | 北京百度网讯科技有限公司 | Navigation guide point method for digging, device, equipment and storage medium |
| CN110399686B (en) * | 2019-07-30 | 2024-03-19 | 中国民航大学 | Parameter-independent aircraft flight trajectory clustering method based on contour coefficients |
| CN110399686A (en) * | 2019-07-30 | 2019-11-01 | 中国民航大学 | A kind of unrelated aircraft flight profiles clustering method of parameter based on silhouette coefficient |
| CN110609824A (en) * | 2019-09-09 | 2019-12-24 | 南京师范大学 | Hot spot area detection method based on dynamic space network model under urban road network environment |
| US11740102B2 (en) | 2019-10-10 | 2023-08-29 | Beijing Baidu Netcom Science Technology Co., Ltd. | Method, apparatus, device and storage medium for determining point of interest area |
| CN110726418A (en) * | 2019-10-10 | 2020-01-24 | 北京百度网讯科技有限公司 | Method, device and equipment for determining interest point region and storage medium |
| CN110849379B (en) * | 2019-10-23 | 2023-04-25 | 南通大学 | Entrance and exit traffic state symbol expression method for navigation map |
| CN110849379A (en) * | 2019-10-23 | 2020-02-28 | 南通大学 | Access and exit traffic state symbol expression method for navigation map |
| CN110866559A (en) * | 2019-11-14 | 2020-03-06 | 上海中信信息发展股份有限公司 | Poultry behavior analysis method and device |
| CN111121803B (en) * | 2019-11-27 | 2021-12-03 | 北京中交兴路信息科技有限公司 | Method and device for acquiring common stop points of road |
| CN111121803A (en) * | 2019-11-27 | 2020-05-08 | 北京中交兴路信息科技有限公司 | Method and device for acquiring common stop points of road |
| CN111275962A (en) * | 2019-12-30 | 2020-06-12 | 深圳市麦谷科技有限公司 | Vehicle track data aggregation effect prediction method and device |
| CN111708853B (en) * | 2020-05-25 | 2022-08-30 | 安徽师范大学 | Taxi hot spot region extraction method based on characteristic density peak clustering |
| CN111708853A (en) * | 2020-05-25 | 2020-09-25 | 安徽师范大学 | Taxi hot spot region extraction method based on characteristic density peak clustering |
| CN114120018A (en) * | 2020-08-25 | 2022-03-01 | 四川大学 | Space activity quantification method based on crowd clustering track entropy |
| CN112163370A (en) * | 2020-09-18 | 2021-01-01 | 国网冀北电力有限公司承德供电公司 | Power distribution network line loss calculation method based on improved ASMDE algorithm |
| CN112070179A (en) * | 2020-09-22 | 2020-12-11 | 中国人民解放军国防科技大学 | Adaptive space-time trajectory clustering method based on density peak value |
| CN112070179B (en) * | 2020-09-22 | 2024-05-31 | 中国人民解放军国防科技大学 | Self-adaptive space-time track clustering method based on density peak value |
| CN113158817A (en) * | 2021-03-29 | 2021-07-23 | 南京信息工程大学 | Objective weather typing method based on rapid density peak clustering |
| CN113158817B (en) * | 2021-03-29 | 2023-07-18 | 南京信息工程大学 | Objective weather typing method based on rapid density peak clustering |
| CN113611115B (en) * | 2021-08-06 | 2022-06-21 | 安徽师范大学 | Vehicle track clustering method based on road network sensitive characteristics |
| CN113611115A (en) * | 2021-08-06 | 2021-11-05 | 安徽师范大学 | Vehicle track clustering method based on road network sensitive characteristics |
| CN113869465A (en) * | 2021-12-06 | 2021-12-31 | 深圳大学 | I-nice algorithm optimization method, device, equipment and computer readable storage medium |
| CN114485696A (en) * | 2021-12-23 | 2022-05-13 | 高德软件有限公司 | Method for acquiring arrival point, electronic device and storage medium |
| CN114677048A (en) * | 2022-04-22 | 2022-06-28 | 北京阿帕科蓝科技有限公司 | Demand area mining method |
| CN114677048B (en) * | 2022-04-22 | 2024-01-16 | 北京阿帕科蓝科技有限公司 | Method for excavating demand area |
| CN114999151A (en) * | 2022-05-24 | 2022-09-02 | 电子科技大学 | Urban traffic flow hierarchical analysis method and device based on density in GPS track |
| CN114999151B (en) * | 2022-05-24 | 2023-01-24 | 电子科技大学 | Density-based urban traffic flow hierarchical analysis method and device in GPS track |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108427965B (en) | 2022-08-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108427965A (en) | A kind of hot spot region method for digging based on road network cluster | |
| CN108629978B (en) | Traffic track prediction method based on high-dimensional road network and recurrent neural network | |
| Chen et al. | Social functional mapping of urban green space using remote sensing and social sensing data | |
| Karagiorgou et al. | On vehicle tracking data-based road network generation | |
| Wang et al. | Automatic intersection and traffic rule detection by mining motor-vehicle GPS trajectories | |
| Clevenger et al. | GIS‐generated, expert‐based models for identifying wildlife habitat linkages and planning mitigation passages | |
| CN103593430B (en) | Clustering method based on mobile object spatiotemporal information trajectory subsections | |
| CN103533501B (en) | A kind of geography fence generation method | |
| CN113378891B (en) | Urban area relation visual analysis method based on track distribution representation | |
| CN105138668A (en) | Urban business center and retailing format concentrated area identification method based on POI data | |
| CN105788274A (en) | Urban intersection lane-level structure extraction method based on time-space trajectory big data | |
| CN108446470A (en) | Medical facilities analysis method of reachability based on track of vehicle data and population distribution | |
| CN106920387A (en) | Obtain the method and device of route temperature in traffic route | |
| Guyot et al. | The urban form of Brussels from the street perspective: The role of vegetation in the definition of the urban fabric | |
| CN113806419B (en) | Urban area function recognition model and recognition method based on space-time big data | |
| CN110413855A (en) | A kind of region entrance Dynamic Extraction method based on taxi drop-off point | |
| Chen et al. | Uvlens: urban village boundary identification and population estimation leveraging open government data | |
| Zhang et al. | Distance Weight-Graph Attention Model-Based High-Resolution Remote Sensing Urban Functional Zone Identification. | |
| CN113342873B (en) | Population analysis unit division method based on city morphology and convergence mode | |
| CN113888867B (en) | Parking space recommendation method and system based on LSTM (least squares) position prediction | |
| CN113903173A (en) | Vehicle track feature extraction method based on directed graph structure and LSTM | |
| Padmanaban | Integrating of urban growth modelling and utility management system using spatio temporal data mining | |
| CN116958649A (en) | Street space city function classification method and related equipment | |
| CN116628078A (en) | Hierarchical urban space structure identification method and system | |
| CN116340563A (en) | Urban scene geographic position positioning method with pattern matching |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |