CN104574965A - City traffic hot spot region partition method based on massive traffic flow data - Google Patents
City traffic hot spot region partition method based on massive traffic flow data Download PDFInfo
- Publication number
- CN104574965A CN104574965A CN201510014926.0A CN201510014926A CN104574965A CN 104574965 A CN104574965 A CN 104574965A CN 201510014926 A CN201510014926 A CN 201510014926A CN 104574965 A CN104574965 A CN 104574965A
- Authority
- CN
- China
- Prior art keywords
- document
- word
- car
- corpus
- 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
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/338—Presentation of query results
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a city traffic hot spot region partition method based on massive traffic flow data. The method comprises the steps that firstly, vehicle passing information data obtained by all external field device checkpoints are sent to a central database, a vehicle passing record file formed by serial numbers of the checkpoints which each vehicle passes by is obtained, models are extracted according to LDA themes, checkpoint vehicle passing information data are transformed into a needed corpus file form, modeling is conducted on the checkpoint vehicle passing information data, theme random sampling is conducted on all words in corpus files until convergence is achieved, and finally the checkpoints with the highest probabilities in all the themes are taken to represent certain traffic hot spot regions. By means of the city traffic hot spot region partition method, the traffic hot spot regions formed by driving route rules of all the vehicles in different time frames can be found, so that analysis bases are provided for the traffic administration department to work out the city traffic jam situations in different time frames, and references are provided for road planning and construction of the city planning department.
Description
Technical field
The invention belongs to data mining technology field, be specifically related to a kind of urban transportation hot spot region division methods based on magnanimity traffic flow data.
Background technology
Along with sustained and rapid development of economy, vehicle guaranteeding organic quantity rapidly increases the traffic problems brought to city and day by day highlights, and congested in traffic, wagon flow freely, greatly not have impact on people's trip speed, and then reduces production and work efficiency.At present, City Traffic Monitor System comparatively perfect, each arterial highway and crossing are all provided with bayonet socket Real-time Collection and cross car data.Cross car data by what gather in analysis city traffic supervisory system, thus find the traffic route of vehicle, utilize the frequent bayonet socket simultaneously occurred that traffic route can be hidden between digging vehicle.With this foundation into dividing urban transportation hot spot region that hides Info, and calculate its heavy traffic index, can for vehicle supervision department provide formulate traffic administration scheme reference, provide the reference etc. of formulating roading for roading department, finally to offer help for transport solution problem.
The traffic route of research vehicle, often can find that many vehicles have similar driving region.Such as, Hangzhou, Zhejiang province city working day early, evening peak, have a large amount of cars and travel on urban district and be communicated with arterial highway with Binjiang District, river-spanning bridge must be these vehicle colonies the urban transportation hot spot region of process; At weekend, must there be each large arterial highway towards scenic spot, the West Lake urban transportation hot spot region of trip.The principal character of above-mentioned urban transportation hot spot region can be summarized as: the different vehicle of a group at close time point through series of identical bayonet socket, and region is just made up of these bayonet sockets, the vehicle fleet passing through these bayonet sockets is designated as the heavy traffic index of this urban transportation hot spot region.Find that the urban transportation hot spot region in the different time sections of city divides and calculates its heavy traffic index, the effective means of blocking up of alleviating can be provided to offer help for vehicle supervision department.
Summary of the invention
The present invention is directed to the deficiencies in the prior art, provide a kind of urban area division methods based on magnanimity traffic flow data.
The central scope of technical solution of the present invention is: utilize the thought extracting theme in natural language processing from the corpus that word and document form to solve the problem of urban transportation hot spot region division; By in a period of time, in traffic surveillance and control system (monitoring, gather, deposit the system of city main trunk road and crossing real-time traffic situation), all car datas of crossing form corpus, wherein each car forms one section of document corresponding to this car at the record that each bayonet socket occurs, and these bayonet sockets numbering is as the word in document, and extract the theme corresponding urban transportation hot spot region of the wherein implicit bayonet socket often simultaneously occurred numbering set as corpus;
The inventive method comprises the following steps:
Step (1). the car information data of crossing obtained by each airfield equipment bayonet socket is sent to central database, crosses car information data and comprises license plate number, spends the car time, crosses car direction and cross car bayonet socket numbering;
Step (2). according to LDA (Latent Dirichlet Allocation) subject distillation model, to obtain in certain setting-up time section all bayonet sockets in central database and cross car information data, be translated into required corpus form and to its modeling, be specially:
Obtain all bayonet socket numberings in city as word composition dictionary, the all bayonet sockets got from central database in certain setting-up time section cross car information data, within this time period, the document of car record excessively formed is numbered by the bayonet socket of process again from wherein obtaining each car, then the car record document of crossing of all vehicles is merged formation corpus document, to add up in this corpus document each car at the probability of occurrence of each bayonet socket, be kept in the probability matrix of a license plate number-bayonet socket, each unit of this matrix represents the probability that certain car occurs at certain bayonet socket;
If total N number of bayonet socket, i.e. N number of word, be designated as: word
j, j=0,1 ..., (N-1), the car occurred in certain setting-up time section has M, i.e. M section document, is designated as: doc
ii=0,1, ..., (M-1), corpus document can be obtained, the bayonet socket numbered sequence of each line item car process within this time period of this corpus document by all car information datas of crossing, the number of times that in statistics corpus document, in dictionary corresponding to every section of document, each word occurs and the total words of every section of document, be designated as: a respectively
i,j, dwSum
i, and by calculating document-word probability matrix, be designated as: M
doc-word, the common M of this matrix is capable, N row, and corresponding i-th car of the i-th row jth row of matrix is through the probability of a jth bayonet socket within this time period, and computing method are:
Carry out the modeling of LDA theme to corpus document, suppose that urban transportation hot spot region number is K, namely K theme, is designated as: topic
k, k=0,1 ..., (K-1), according to LDA model, suppose that in corpus document, document and theme meet the Dirichlet distribute being Study first with α and β respectively, so in corpus, the generative process of every section of document and each theme meets respectively
with
multinomial distribution, then in corpus, the distribution of document-theme and the distribution of theme-word can be expressed as doc-topic probability matrix, topic-word probability matrix, are designated as: M
doc-topic, M
topic-word, final M
topic-wordbe the model result of LDA;
Step (3). utilize Gibbs Sampling algorithm, theme stochastic sampling is carried out to words all in corpus until Gibbs Sampling convergence obtains M
topic-word, M
topic-wordrow k represent all word
j(i.e. bayonet socket numbering) is at this topic
kunder probability distribution, get each topic
kc the word that middle probability is the highest
j(i.e. bayonet socket numbering) represents certain traffic hot spot region (c is determined by the size in required division traffic hot spot region);
Step (4). the total car amount excessively corresponding to the bayonet socket numbering set in the traffic hot spot region that statistic procedure (3) obtains is designated as the heavy traffic index in this traffic hot spot region.
A kind of urban transportation hot spot region division methods based on magnanimity traffic flow data provided by the present invention is made up of one group of functional module, and they comprise: cross car information data acquisition module, cross car information data transform and LDA theme MBM, Gibbs Sampling theme sampling module and traffic hot spot region heavy traffic index computing module.
That crosses that car information data acquisition module obtains each airfield equipment bayonet socket Real-time Collection crosses car information data, comprises car license plate number, and spent the car time, crosses car direction and crosses car bayonet socket numbering, and be saved to central database.
Cross car information data transform and LDA theme MBM to cross car information data for raw data, be translated into the form needed for the modeling of LDA theme and complete modeling, with M
topic-wordfor modeling result.
Gibbs Sampling theme sampling module on the basis of LDA theme modeling, to M
topic-wordin each word
jadopt Gibbs Sampling algorithm to carry out theme sampling until convergence, obtain each topic
kcorresponding bayonet socket numbering set.
The heavy traffic index computing module in traffic hot spot region mainly adds up each topic
kthe car total amount of crossing of the corresponding bayonet socket of corresponding bayonet socket numbering set is designated as this topic
kthe heavy traffic index in the traffic hot spot region represented.
The beneficial effect that the present invention has:
The present invention utilizes the magnanimity of City Traffic Monitor System collection to cross car information data, finds the region characteristic that the traffic route rule of all vehicles in different time sections is formed, as the foundation dividing hot spot region, city.This result can provide analysis foundation for vehicle supervision department solves urban traffic blocking situation in different time sections, for urban planning authority's roading, builds and provides reference, for resident's trip of driving provides induction.
On the basis that the present invention divides in urban transportation hot spot region, the busy index of each urban transportation hot spot region can be added up.This busy index can reflect city focus regional traffic active degree to a certain extent.
Accompanying drawing explanation
Fig. 1 dictionary schematic diagram;
Fig. 2 corpus document schematic diagram;
Fig. 3 license plate number-bayonet socket numbering probability matrix (i.e. document-word frequencies matrix) schematic diagram.
Embodiment
Below in conjunction with accompanying drawing, the invention will be further described.
Step (1). the car information data crossed obtained by each airfield equipment bayonet socket is sent to central database, crosses car information data and comprises: license plate number, spend the car time, cross car direction and cross car bayonet socket numbering.
Step (2). according to LDA (Latent Dirichlet Allocation) subject distillation model, to obtain in certain setting-up time section all bayonet sockets in central database and cross car information data, be translated into required corpus form and to its modeling, be specially:
Obtain all bayonet socket numberings in city as word composition dictionary (see Fig. 1), the all bayonet sockets got from central database in certain setting-up time section cross car information data, then within this time period, number the car record document of crossing form by the bayonet socket of process from wherein obtaining each car (i-th car is represented is document doc
i, in document, the bayonet socket numbering of the n-th process is expressed as word word
i,s, s=0,1 ..., (n
i-1), n
ibe the bayonet socket numbering number of i-th car process, word
i,s∈ { word
j| j=0,1, ..., (N-1) }), then the car record document of crossing of all vehicles is merged formation corpus document (see Fig. 2), adding up each car in this corpus document, at the probability of occurrence of each bayonet socket, is kept in the probability matrix of a license plate number-bayonet socket, and each unit of this matrix represents the probability that certain car occurs at certain bayonet socket.
If total N number of bayonet socket, i.e. N number of word, be designated as: word
j, j=0,1 ..., (N-1), the car occurred in certain setting-up time section has M, i.e. M section document, is designated as: doc
ii=0,1, ..., (M-1), corpus document can be obtained, the bayonet socket numbered sequence of each line item car process within this time period of this corpus document by all car information datas of crossing, statistics corpus document obtains the probability matrix of license plate number-bayonet socket (i.e. document-word), is designated as: M
doc-word(see Fig. 3, the common M of this matrix is capable, N row, and corresponding i-th car of the i-th row jth row of matrix is through the probability of a jth bayonet socket within this time period, and the probability sum of the every a line of matrix is 1).
Carry out the modeling of LDA theme to corpus document, suppose that urban transportation hot spot region number is K, namely K theme, is designated as: topic
k, k=0,1 ..., (K-1), according to LDA model, suppose that in corpus document, document and theme meet the Dirichlet distribute being Study first with α and β respectively, so in corpus, the generative process of every section of document and each theme meets respectively
with
multinomial distribution, then in corpus, the distribution of document-theme and the distribution of theme-word can be expressed as doc-topic probability matrix, topic-word probability matrix, are designated as: M
doc-topic, M
topic-word, final M
topic-wordbe the model result of LDA.
Step (3). utilize Gibbs Sampling algorithm, theme stochastic sampling is carried out to words all in corpus until Gibbs Sampling convergence obtains M
topic-word, M
topic-wordrow k represent all word
j(i.e. bayonet socket numbering) is at this topic
kunder probability distribution, get each topic
kc the word that middle probability is the highest
j(i.e. bayonet socket numbering) represents certain traffic hot spot region (c is determined by the size in required division traffic hot spot region), concrete steps:
(1) random initializtion: to each word word in every section of document in corpus
j, compose a topic numbering k randomly, i.e. the corresponding topic of this word
k, add up each topic in every section of document
keach word word of dictionary corresponding to corresponding word number and corpus document
jbelong to different topic
knumber be recorded in SM respectively
nmk, SM
nktin matrix.SM
nmkmatrix is capable, the K row of M altogether, wherein in the i-th row kth row record i-th section of document, belong to the word number of a kth theme; SM
nktmatrix is capable, the N row of K altogether, and wherein row k jth row record word j represents the number of times that theme k occurs.And add up the total words of every section of document in corpus and the total words of each theme, be recorded in M respectively
nmkSum, M
nktSumin matrix, M
nmkSummatrix altogether M capable 1 arranges, the i-th row the 1st row record i-th section of document subject matter sum, M
nktSummatrix altogether K capable 1 arranges, and line K the 1st arranges total words corresponding to a record kth theme.
(2) multiple scanning corpus document, every section of document doc
iin corresponding each word word
jtopic numbering k, obtain φ [k] [j], ζ [i] [k] according to Gibbs Sampling formula:
(α, β press empirical value setting), with K
value adopts the multinomial method of sampling, resampling doc
imiddle word
jtopic numbering, by sampling result upgrade SM
nmk, SM
nkt, M
nmkSumand M
nktSummatrix.
(3) sampling process n time of (2) is repeated until Gibbs Sampling restrains (n presses empirical value setting).
(4) M is calculated
nmk, M
nktin every value, computing method are:
m
nktnamely the result of model, gets M
nktword corresponding to every a line c maximal value
jas this row topic
kthe bayonet socket numbering set of representative.
Step (4). the total car amount excessively corresponding to the bayonet socket numbering set in the traffic hot spot region that statistic procedure (3) obtains, is designated as the heavy traffic index in this traffic hot spot region.
Claims (1)
1., based on a urban transportation hot spot region division methods for magnanimity traffic flow data, it is characterized in that the concrete steps of the method are:
Step (1). the car information data crossed obtained by each airfield equipment bayonet socket is sent to central database, crosses car information data and comprises: license plate number, spend the car time, cross car direction and cross car bayonet socket numbering;
Step (2). according to LDA subject distillation model, to obtain in certain setting-up time section all bayonet sockets in central database and cross car information data, be translated into required corpus form and to its modeling, be specially:
Obtain all bayonet socket numberings in city as word composition dictionary, the all bayonet sockets got from central database in certain setting-up time section cross car information data, within this time period, the document of car record excessively formed is numbered by the bayonet socket of process again from wherein obtaining each car, then the car record document of crossing of all vehicles is merged formation corpus document, to add up in this corpus document each car at the probability of occurrence of each bayonet socket, be kept in the probability matrix of a license plate number-bayonet socket, each unit of this matrix represents the probability that certain car occurs at certain bayonet socket;
If total N number of bayonet socket, i.e. N number of word, be designated as: word
j, j=0,1 ..., (N-1), the car occurred in certain setting-up time section has M, i.e. M section document, is designated as: doc
ii=0,1, ..., (M-1), corpus document can be obtained, the bayonet socket numbered sequence of each line item car process within this time period of this corpus document by all car information datas of crossing, the number of times that in statistics corpus document, in dictionary corresponding to every section of document, each word occurs and the total words of every section of document, be designated as: a respectively
i,j, dwSum
i; And by calculating document-word probability matrix, be designated as: M
doc-word, the common M of this matrix is capable, N row, and corresponding i-th car of the i-th row jth row of matrix is through the probability of a jth bayonet socket within this time period, and this value is:
Carry out the modeling of LDA theme to corpus document, suppose that urban transportation hot spot region number is K, namely K theme, is designated as: topic
k, k=0,1 ..., (K-1), according to LDA model, suppose that in corpus document, document and theme meet the Dirichlet distribute being Study first with α and β respectively, so in corpus, the generative process of every section of document and each theme meets respectively
with
multinomial distribution, then in corpus, the distribution of document-theme and the distribution of theme-word can be expressed as doc-topic probability matrix, topic-word probability matrix, are designated as: M
doc-topic, M
topic-word, final M
topic-wordbe the model result of LDA;
Step (3). utilize Gibbs Sampling algorithm, theme stochastic sampling is carried out to words all in corpus until Gibbs Sampling convergence obtains M
topic-word, M
topic-wordrow k represent all word
jat this topic
kunder probability distribution, get each topic
kc the word that middle probability is the highest
jrepresent certain traffic hot spot region, c is determined by the size in required division traffic hot spot region;
Step (4). the total car amount excessively corresponding to the bayonet socket numbering set in the traffic hot spot region that statistic procedure (3) obtains is designated as the heavy traffic index in this traffic hot spot region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510014926.0A CN104574965B (en) | 2015-01-11 | 2015-01-11 | A kind of urban transportation hot spot region based on magnanimity traffic flow data division methods |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510014926.0A CN104574965B (en) | 2015-01-11 | 2015-01-11 | A kind of urban transportation hot spot region based on magnanimity traffic flow data division methods |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104574965A true CN104574965A (en) | 2015-04-29 |
| CN104574965B CN104574965B (en) | 2017-01-04 |
Family
ID=53090908
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510014926.0A Active CN104574965B (en) | 2015-01-11 | 2015-01-11 | A kind of urban transportation hot spot region based on magnanimity traffic flow data division methods |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104574965B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809967A (en) * | 2016-05-25 | 2016-07-27 | 浙江宇视科技有限公司 | Traffic flow displaying method and device |
| CN106095976A (en) * | 2016-06-20 | 2016-11-09 | 杭州电子科技大学 | A kind of interest Dimensional level extracting method based on microblog data supporting OLAP to apply |
| CN106611499A (en) * | 2015-10-21 | 2017-05-03 | 北京计算机技术及应用研究所 | Method of detecting vehicle hotspot path |
| CN106709595A (en) * | 2016-11-24 | 2017-05-24 | 北京交通大学 | Accident delay time prediction method and system based on unformatted information |
| CN107862862A (en) * | 2016-09-22 | 2018-03-30 | 杭州海康威视数字技术股份有限公司 | A kind of vehicle behavior analysis method and device |
| CN111126713A (en) * | 2019-12-31 | 2020-05-08 | 方正国际软件(北京)有限公司 | Space-time hot spot prediction method and device based on bayonet data and controller |
| CN111932873A (en) * | 2020-07-21 | 2020-11-13 | 重庆交通大学 | Real-time traffic early warning management and control method and system for mountain city hot spot area |
| CN112950932A (en) * | 2021-01-26 | 2021-06-11 | 阿里巴巴集团控股有限公司 | Method and device for merging area units and electronic equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120041953A1 (en) * | 2010-08-16 | 2012-02-16 | Microsoft Corporation | Text mining of microblogs using latent topic labels |
| CN103544841A (en) * | 2013-10-22 | 2014-01-29 | 银江股份有限公司 | Method for dynamically dividing traffic control region |
| CN103678277A (en) * | 2013-12-04 | 2014-03-26 | 东软集团股份有限公司 | Theme-vocabulary distribution establishing method and system based on document segmenting |
| KR20140065235A (en) * | 2012-11-21 | 2014-05-29 | 서울대학교산학협력단 | Apparatus and method for traffic classificaiton |
| CN104050298A (en) * | 2014-07-04 | 2014-09-17 | 成都品果科技有限公司 | User segmentation method based on LDA topic model algorithm |
| KR20140114496A (en) * | 2013-03-14 | 2014-09-29 | 한국과학기술원 | Method and system for providing summery of text document using word cloud |
| CN104156350A (en) * | 2014-08-04 | 2014-11-19 | 苏州大学 | Test semantic extraction method based on fine division of MapReduce |
-
2015
- 2015-01-11 CN CN201510014926.0A patent/CN104574965B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120041953A1 (en) * | 2010-08-16 | 2012-02-16 | Microsoft Corporation | Text mining of microblogs using latent topic labels |
| KR20140065235A (en) * | 2012-11-21 | 2014-05-29 | 서울대학교산학협력단 | Apparatus and method for traffic classificaiton |
| KR20140114496A (en) * | 2013-03-14 | 2014-09-29 | 한국과학기술원 | Method and system for providing summery of text document using word cloud |
| CN103544841A (en) * | 2013-10-22 | 2014-01-29 | 银江股份有限公司 | Method for dynamically dividing traffic control region |
| CN103678277A (en) * | 2013-12-04 | 2014-03-26 | 东软集团股份有限公司 | Theme-vocabulary distribution establishing method and system based on document segmenting |
| CN104050298A (en) * | 2014-07-04 | 2014-09-17 | 成都品果科技有限公司 | User segmentation method based on LDA topic model algorithm |
| CN104156350A (en) * | 2014-08-04 | 2014-11-19 | 苏州大学 | Test semantic extraction method based on fine division of MapReduce |
Non-Patent Citations (4)
| Title |
|---|
| 李庆丰: "基于主题模型的多文档自动文摘方法研究", 《中国优秀硕士学位论文全文数据库（电子期刊）》 * |
| 王振振，等: "基于ＬＤＡ主题模型的文本相似度计算", 《计算机科学》 * |
| 石晶，等: "基于LDA模型的主题分析", 《自动化学报》 * |
| 石晶，等: "基于LDA模型的文本分割", 《计算机学报》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106611499A (en) * | 2015-10-21 | 2017-05-03 | 北京计算机技术及应用研究所 | Method of detecting vehicle hotspot path |
| CN105809967A (en) * | 2016-05-25 | 2016-07-27 | 浙江宇视科技有限公司 | Traffic flow displaying method and device |
| CN106095976A (en) * | 2016-06-20 | 2016-11-09 | 杭州电子科技大学 | A kind of interest Dimensional level extracting method based on microblog data supporting OLAP to apply |
| CN106095976B (en) * | 2016-06-20 | 2019-09-24 | 杭州电子科技大学 | A kind of interest Dimensional level extracting method based on microblog data for supporting OLAP to apply |
| CN107862862A (en) * | 2016-09-22 | 2018-03-30 | 杭州海康威视数字技术股份有限公司 | A kind of vehicle behavior analysis method and device |
| CN107862862B (en) * | 2016-09-22 | 2020-11-20 | 杭州海康威视数字技术股份有限公司 | Vehicle behavior analysis method and device |
| CN106709595A (en) * | 2016-11-24 | 2017-05-24 | 北京交通大学 | Accident delay time prediction method and system based on unformatted information |
| CN111126713A (en) * | 2019-12-31 | 2020-05-08 | 方正国际软件(北京)有限公司 | Space-time hot spot prediction method and device based on bayonet data and controller |
| CN111126713B (en) * | 2019-12-31 | 2023-05-09 | 方正国际软件(北京)有限公司 | Space-time hot spot prediction method and device based on bayonet data and controller |
| CN111932873A (en) * | 2020-07-21 | 2020-11-13 | 重庆交通大学 | Real-time traffic early warning management and control method and system for mountain city hot spot area |
| CN111932873B (en) * | 2020-07-21 | 2022-10-04 | 重庆交通大学 | Real-time traffic early warning management and control method and system for mountain city hot spot area |
| CN112950932A (en) * | 2021-01-26 | 2021-06-11 | 阿里巴巴集团控股有限公司 | Method and device for merging area units and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104574965B (en) | 2017-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104574965A (en) | City traffic hot spot region partition method based on massive traffic flow data | |
| Arcaute et al. | Cities and regions in Britain through hierarchical percolation | |
| CN107610469A (en) | A kind of day dimension regional traffic index forecasting method for considering multifactor impact | |
| Ma et al. | A novel STFSA-CNN-GRU hybrid model for short-term traffic speed prediction | |
| CN111161535A (en) | Attention mechanism-based graph neural network traffic flow prediction method and system | |
| Liu et al. | Spatiotemporal variation analysis of regional flood disaster resilience capability using an improved projection pursuit model based on the wind-driven optimization algorithm | |
| CN102610092A (en) | Urban road speed predication method based on RBF (radial basis function) neural network | |
| Lin et al. | A spatial-temporal hybrid model for short-term traffic prediction | |
| CN105095993A (en) | System and method for predicting passenger flow volume of railway stations | |
| CN105183841A (en) | Recommendation method in combination with frequent item set and deep learning under big data environment | |
| Mahmoud et al. | Spatiotemporal variation analysis of urban land expansion in the establishment of new communities in Upper Egypt: A case study of New Asyut city | |
| CN105975582B (en) | RFID data based on Spark generates the method and system of travelling OD matrix | |
| CN103903429A (en) | Fast bus stop delay time combination predicting method | |
| CN105070097A (en) | Risk recognition method for operating vehicle road traffic accidents | |
| Min et al. | Traffic flow data recovery algorithm based on gray residual GM (1, N) model | |
| CN111144281B (en) | Urban rail transit OD passenger flow estimation method based on machine learning | |
| Lu et al. | Analyzing the evolution of rare events via social media data and k-means clustering algorithm | |
| CN103914981A (en) | Method for predicting confliction between pedestrians and left-turn vehicles at plane intersection | |
| Huang et al. | Spatiotemporal evolution of land transportation networks and accessibility in inland mountainous areas 1917–2017: A case study of Southwest China | |
| CN102880753B (en) | Based on the land utilization space characteristic dimension conversion method of fractal dimension | |
| Du et al. | Highway travel time prediction of segments based on ANPR data considering traffic diversion | |
| Jia et al. | Traffic flow prediction based on self-attention mechanism and deep packet residual network | |
| Yang et al. | Network traffic prediction based on LSSVM optimized by PSO | |
| Wang | Road condition perception model based on adaptive dynamic path planning algorithm and time series. | |
| Tiwari et al. | Predicting Future Urban Growth in Bhopal, India using Artificial Intelligence based Modelling |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20150429 Assignee: Hangzhou Cheng Road Polytron Technologies Inc Assignor: Hangzhou Electronic Science and Technology Univ Contract record no.: 2019330000034 Denomination of invention: City traffic hot spot region partition method based on massive traffic flow data Granted publication date: 20170104 License type: Common License Record date: 20190319 |
|
| EE01 | Entry into force of recordation of patent licensing contract |