CN104112366A - Method for traffic signal optimization based on latent semantic model - Google Patents
Method for traffic signal optimization based on latent semantic model Download PDFInfo
- Publication number
- CN104112366A CN104112366A CN201410360312.3A CN201410360312A CN104112366A CN 104112366 A CN104112366 A CN 104112366A CN 201410360312 A CN201410360312 A CN 201410360312A CN 104112366 A CN104112366 A CN 104112366A
- Authority
- CN
- China
- Prior art keywords
- timing scheme
- traffic
- scoring
- traffic behavior
- model
- 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
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000005457 optimization Methods 0.000 title claims abstract description 23
- 238000012549 training Methods 0.000 claims abstract description 17
- 230000006399 behavior Effects 0.000 claims description 55
- 238000012360 testing method Methods 0.000 claims description 17
- 238000010801 machine learning Methods 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 8
- 239000004576 sand Substances 0.000 claims description 4
- 238000002474 experimental method Methods 0.000 claims description 3
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 238000011217 control strategy Methods 0.000 abstract description 4
- 238000012545 processing Methods 0.000 abstract description 2
- 238000013507 mapping Methods 0.000 abstract 1
- 238000007781 pre-processing Methods 0.000 abstract 1
- 230000009469 supplementation Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 6
- 238000011160 research Methods 0.000 description 5
- 239000013598 vector Substances 0.000 description 5
- 238000007418 data mining Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 230000006698 induction Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000000819 phase cycle Methods 0.000 description 2
- 230000001502 supplementing effect Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Abstract
The invention discloses a method for traffic signal optimization based on a latent semantic model. According to the method, the advantages of the latent semantic model in processing uncertain factors in a recommendation system are utilized, a traffic state is subject to analog to be a user, timing plans are subject to analog to be articles, and traffic indexes (such as time delay) are subject to analog to be scores. The method comprises the steps of preprocessing data; mapping the traffic state and the timing plans to a latent semantic space; training a score prediction model of the traffic state for the timing plans; predicting the scores and selecting an optimal timing plan; and practically applying the timing plan and feeding back traffic information. The method is simple and practical, the uncertain factors which cannot be treated well through a traditional control strategy and have large influence in practice can be treated, the uncertain factors which cannot be accurately quantized easily do not need to undergo modeling accurately, and the method has very good universality and less constraints and can serve as supplementation and optimization of the traditional strategy.
Description
Technical field
The present invention relates to the technical fields such as traffic signals control, information processing and data mining, more specifically, relate to a kind of traffic signal optimization method based on hidden semantic model.
Background technology
The continuous growth of Urban vehicles poputation, causes a lot of urban issueses, as traffic congestion, environmental pollution, traffic hazard etc.Depend on merely to increase and build traffic infrastructure, as expanded road etc., both consumed wealth larger, need again in the face of realistic problems such as the existing planning of building removal and city.Traffic signals are controlled the mode that is proved to be in practice a kind of effective alleviation traffic problems, mainly by providing real-time timing scheme to realize to various road conditions complicated and changeable.Traffic signals in theoretical research are controlled, attempt to analyze the various factors that may affect traffic in actual traffic environment, by idealized or hypothetical modeling, with mathematical linguistics, the traffic environment of closing to reality is as far as possible described, further predict the road conditions of next period, to configure in advance suitable traffic lights timing scheme, to alleviate as much as possible traffic congestion.Based on theoretical research, current traffic signal control strategy can roughly be divided into timing controlled, induction control, Based Intelligent Control etc.; But be limited to the complicacy of actual traffic environment and randomness etc., in practical application, take elementary Time controlling schema as main.And cause the principal element of its complicacy, randomness etc., be exactly in actual traffic environment, to have a large amount of uncertain factors, these factors are difficult to mathematical linguistics accurate modeling, even have quite a few uncertain factor cannot modeling, and this just causes theoretic idealized hypothesis cannot reach gratifying degree in actual applications.
Commending system is mainly used in the fields such as ecommerce, web film, music site, video website, advertisement at present; Personalized recommendation system, by setting up the binary relation between user and information products, utilizes the potential interested object of existing selection course or each user of similarity relation excavation, and then carries out personalized recommendation.And hidden semantic model (Latent Factor Model) is the most popular research topic in commending system field in recent years, its core concept is by hidden feature (latent factor) contact user interest and article; Hidden semantic model is obtained good effect and is widely used, and its main cause is that it can process interactional a lot of uncertain factors, i.e. hidden feature between user interest and article preferably.Hidden semantic model is attempted the scoring to article by analysis user, and user and article are mapped to a hidden semantic space; Article and user are by a vector representation, and vector element is various uncertain factors.For example, take film as example, some factor is dominant, as comedy, action movie, horror film etc., some factor is very difficult well-defined, as the moral degree of depth, strange degree etc., also has several factors just cannot explain at all, but these factors are all to affect the fancy grade of user to film, and influencing each other between user and film played to critical effect.The degree of these elements of user preferences is higher, and film to have the degree of these elements higher, user more easily likes these films.Hidden semantic model, from the angle of machine learning and data mining, well processed the uncertain factor with mathematics Accurate Model that is difficult between user and article, well set up being connected of optimum matching of user and article.
Increasingly mature along with detection technique, the data volume in traffic also grows with each passing day, and makes applied for machines study and data mining technology in traffic become possibility; And in traffic, also exist some indexs can reflect " fancy grade " of traffic behavior to traffic signals, as the time delay of traffic behavior under corresponding Traffic Signal Timing scheme, flow etc.How by being applied to during traffic signals recommend of hidden semantic model seamless link, handling well and in complicated traffic environment, be difficult to modeling but to the influential uncertain factor of actual traffic situation, be a research point of challenging greatly and have meaning.
Summary of the invention
For in current actual traffic environment, exist be difficult in a large number with mathematics Accurate Model, even at all cannot modeling uncertain factor, and these factors cause the gap problem between theoretical research result and practical application, the object of this invention is to provide a kind of traffic signal optimization method based on hidden semantic model, to solve the seamless link problem of theoretical result and practical application.
To achieve these goals, the present invention proposes a kind of traffic signal optimization method based on hidden semantic model, comprise the following steps:
Selected relevant traffic indicators is scoring modeling;
For basic traffic behavior and the modeling of timing scheme;
In conjunction with actual traffic situation, consider traffic behavior and timing scheme self character, further the accurate described score in predicting model of refinement;
Utilize data and the optimization method of database, from the parameter of score in predicting model described in the angle exercise of machine learning, to obtain the score in predicting parameters of formula of traffic behavior to timing scheme;
The described score in predicting parameters of formula that utilization obtains, in conjunction with described score in predicting model, the scoring of prediction traffic behavior to original timing scheme;
The historical used timing scheme of the described best timing scheme that contrast obtains and this traffic behavior, draws final optimal timing scheme.
Wherein, the data acquisition of described database both can gather by technology, also can obtain by Traffic Administration Bureau; And the timing scheme in described database both can, by collecting, can be generated by classic algorithm again.
Wherein, as the traffic indicators of scoring, choose time delay as scoring.
Wherein, the described step for basic traffic behavior and the modeling of timing scheme comprises: traffic behavior and timing scheme are mapped to hidden semantic space with vectorial form, with mathematical linguistics, are expressed as:
q
t∈R
f，p
s∈R
f，
Wherein, vectorial q
trepresent timing scheme, vectorial p
srepresent traffic behavior, f represents interactional uncertain factor between traffic behavior and timing scheme.
Wherein, the step of the accurate described score in predicting model of described further refinement comprises:
With the biasing scoring existing in equation expression traffic behavior and timing scheme, be:
b
st＝u+b
s+b
t，
Thus, final score in predicting model is
Wherein,
for prediction appraisal result, b
strepresent biasing scoring, u represents average score, b
sand b
trepresent respectively traffic behavior and timing scheme marking itself or departed from the degree that draw is divided u, vectorial q by marking
trepresent timing scheme, vectorial p
srepresent traffic behavior.
Wherein, described data and the optimization method that utilizes database, from the parameter of score in predicting model described in the angle exercise of machine learning, further comprises the step of the score in predicting parameters of formula of timing scheme to obtain traffic behavior:
Arrangement extract traffic behavior, timing scheme and corresponding scoring in database all three-number set;
Data pre-service: the data of above-mentioned three-number set are processed into unified standard form on request;
The data of handling well are divided into M part, and wherein M-1 part, as training set, remains 1 part as test set;
Training pattern parameter on training set, selecting to minimize prediction scoring and true absolute mean square deviation of marking is herein target training,
Wherein, r
stfor true scoring, α represents the set of the tlv triple that traffic behavior on training set, timing scheme and corresponding scoring all exist, and in order to prevent the regularization term of over-fitting, λ is regularization parameter to second of above formula, and its occurrence is general to be chosen in conjunction with experiment; By random gradient descent method or least square method, be optimized herein, obtain respectively parameter to be asked;
Obtain after model, on test set, test the absolute mean square deviation under this model, by following formula, calculate:
Wherein, | α | represent the length of the set of the tlv triple that on test set, traffic behavior, timing scheme and corresponding scoring all exist;
Adjust parameter f, repeat above-mentioned two steps, train various model, and the absolute mean square deviation on corresponding test set;
According to above result, select to make the optimum prediction model of absolute mean square deviation RMSE minimum.
Wherein, the step of the described best timing scheme that described contrast obtains and the historical used timing scheme of this traffic behavior comprises: utilize optimum prediction model, for current traffic behavior is predicted the scoring of its original timing scheme, and by the scoring contrast of this scoring timing scheme used with it, determine final best timing scheme.
Preferably, after being also included in and obtaining best timing scheme, further by the data feedback after practical application, give described database, further abundant or upgrade the step of described database.
The present invention compares with classic method, during uncertain factor in processing actual traffic environment, traditional traffic signal control method is faced uncertain factor complicated and changeable in traffic and is had two large problems, the one, the factor that can analyze has been carried out to idealized hypothesis, idealized modeling, the 2nd, to the uncertain factor that cannot analyze or easily ignore, when modeling, selectivity is ignored.The present invention is based upon on the basis of large data, from the framework of machine learning, considers, the result of take is excavated the various uncertain factors in actual traffic environment as leading, and in forecast model, its impact is taken into account.And method of the present invention is simple, be easy to realize, and the reality of dealing with problems, can be used as supplementing and optimizing of conventional traffic signal control strategy.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of the traffic signal optimization method based on hidden semantic model of the present invention;
Fig. 2 is a kind of model schematic diagram of common crossing;
Fig. 3 is a kind of phase sequence figure of common timing scheme;
Fig. 4 is a kind of model schematic diagram of crossing in particular cases;
Fig. 5 is the model schematic diagram at another kind of crossing in particular cases;
Fig. 6 is the process flow diagram of model training of the present invention and system of selection.
Embodiment
For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with specific embodiment, and with reference to accompanying drawing, the present invention is described in further detail.
The method that proposed for a better understanding of the present invention, what a traffic information database we supposition set up by video detection, ground induction coil and other correlation techniques, this database comprises traffic information and corresponding timing scheme, and the time delay under corresponding timing scheme etc.The expansion of following steps, all foundation based on this database.
Specific implementation principle of the present invention is: first, need utilize video detection technology, ground induction coil technology etc. to obtain relevant traffic data, set up a huge database; This database comprises data on flows, incurs loss through delay the data that data, queue length etc. can be used for describing traffic and index, and this database also comprises the timing scheme that traffic state data is corresponding.Be based upon on the basis of so huge database, the present invention can utilize machine learning and data mining technology modeling; While realizing hidden semantic model in traffic environment, traffic behavior is modeled as to user, and timing scheme is modeled as article, and various indexs are modeled as scoring, utilize matrix decomposition technology to realize hidden semantic model, handle well in actual traffic environment in a large number cannot accurate modeling uncertain factor.More specifically, the present invention is by hidden semantic model, according to real-time road to " scoring " of traffic signals (timing scheme) (index such as time delay), real-time road and timing scheme are modeled as to a vector that uncertain factor forms in various traffic, are mapped to a hidden semantic space; Element in this hidden semantic space does not need to be known or to be described, and the matching degree of real-time road and timing scheme is weighed by vector product.
Fig. 1 is the process flow diagram that the present invention is based on the traffic signal optimization method of hidden semantic model, and as shown in Figure 1, the described traffic signal optimization method based on hidden semantic model specifically comprises the following steps:
Step S1 chooses suitable index from database, is scoring modeling.
For Score index, to reflect the appropriate level of timing scheme to current traffic, can choose time delay (delay) as scoring.Because scoring is more high better, and delay is more low better, so after selected delay, carried out data back pre-service, asks the value of 1000/delay.
Step S2 is basic traffic behavior and the modeling of timing scheme.
Traffic behavior is by certain magnitude of traffic flow, to act on intersection to be formed, and the magnitude of traffic flow and road structure influence each other.The most basic a kind of road model as shown in Figure 2, four intersections, every road is three tracks.
Fig. 3 is a kind of typical four intersections phase sequence.Wherein, the green time of four phase places has to be determined, and each phase place is common forms a signal controlling cycle, and in the ordinary course of things, four roads that we give tacit consent to connection intersection always allow right lateral.
Traffic behavior and timing scheme influence each other, and according to hidden semantic model and above-mentioned substantially common traffic behavior and timing scheme, can be modeled as two vectors of hidden semantic space:
q
t∈R
f，p
s∈R
f，
Wherein, vectorial q
trepresent timing scheme, vectorial p
srepresent traffic behavior, f represents the dimension of interactional uncertain factor between traffic behavior and timing scheme.
Be based upon on this model the prediction scoring of traffic behavior to timing scheme
for
Step S3, considers various situations in reality, adds biasing impact, further accurate model.
Be based upon on the basis of step S2, real road and timing method also have various ubiquitous situations, as shown in Figure 4, Figure 5.
Figure 4 shows that a kind of special case of substantially common road structure shown in Fig. 2, A place road is narrower than other three, and this situation causes this road itself just easily to get congestion.
Figure 5 shows that a kind of special case of substantially common road surrounding enviroment shown in Fig. 2, B place road has the units such as school, hospital, generally on these unit doorways, has independent traffic lights, and it is smooth and easy that this situation is not easy this crossing traffic.
Except above two kinds of special cases, also have other multiple situations, as road is very wide, cause itself being just not easy traffic congestion etc.These situations also can affect last scoring, it is the biasing of traffic behavior and timing scheme scoring itself, do not belong to the interaction between traffic behavior and timing scheme, so take into account traffic behavior and the tendentiousness of timing scheme to scoring itself, its scoring can be modeled as
b
st＝u+b
s+b
t，
Wherein, b
strepresent biasing scoring, u represents average score, b
sand b
trepresent respectively traffic behavior and timing scheme marking itself or departed from by marking the degree that draw is divided u.
Integrating step S2 and bias model, final score in predicting model is
Step S4, utilizes data and the optimization method of database, from the model parameter of the angle exercise step S3 of machine learning, and to obtain the score in predicting parameters of formula of traffic behavior to timing scheme, i.e. u, b
sand b
tvalue, and the q under various dimension
tand p
svalue.
As shown in Figure 6, described step S4 is further comprising the steps:
Step S41, arrange traffic behavior, timing scheme and the corresponding scoring extract in database all three-number set;
Step S42, data pre-service: the data obtained in step S41 is processed into unified standard form on request, as for time delay, can unifies to carry out 1000/delay operation;
Step S43, is divided into M part by the data of handling well, and wherein M-1 part, as training set, remains 1 part as test set;
Step S44, training pattern parameter on training set, selecting to minimize prediction scoring and true absolute mean square deviation of marking is herein target training,
Wherein, r
stfor true scoring, α represents the set of the tlv triple that traffic behavior on training set, timing scheme and corresponding scoring all exist, and in order to prevent the regularization term of over-fitting, λ is regularization parameter to second of above formula, and its occurrence is general to be chosen in conjunction with experiment; By random gradient descent method or least square method etc., be optimized herein, obtain respectively parameter to be asked.
Step S45, obtains after model, tests the absolute mean square deviation (root-mean-square error, RMSE) under this model on test set, by following formula, calculates:
Herein | α | represent the length of the set of the tlv triple that on test set, traffic behavior, timing scheme and corresponding scoring all exist.
Step S46, adjusts parameter (enigmatic language justice number f in hidden semantic model), and repeating step S44 and step S45, train various model, and the absolute mean square deviation on corresponding test set.
Step S47, according to above result, selects best forecast model, makes the model of absolute mean square deviation RMSE minimum in step S45.
Step S5, utilizes the optimum model parameter obtaining in step S4, the score in predicting model of integrating step S3
the scoring of prediction traffic behavior to original timing scheme.
Step S6, original scoring in the scoring that integrating step S5 obtains and database, final most suitable timing scheme is determined in contrast.
Step S7, is applied to reality by the timing scheme of step S6, and the testing result after practical application is fed back to database, upgrades or further enriches.
By above-mentioned known to introducing of the inventive method, the present invention is based upon on the basis of large data, from the framework of machine learning, consider, the result of take is excavated the various uncertain factors in actual traffic environment as leading, and in forecast model, its impact is taken into account.By practice test, method of the present invention is simple, can be used as supplementing and optimizing of conventional traffic signal control strategy.
Above-described specific embodiment; object of the present invention, technical scheme and beneficial effect are further described; be understood that; the foregoing is only specific embodiments of the invention; be not limited to the present invention; within the spirit and principles in the present invention all, any modification of making, be equal to replacement, improvement etc., within all should being included in protection scope of the present invention.
Claims (8)
1. the traffic signal optimization method based on hidden semantic model, comprises the following steps:
Selected relevant traffic indicators is scoring modeling;
For basic traffic behavior and the modeling of timing scheme;
In conjunction with actual traffic situation, consider traffic behavior and timing scheme self character, further the accurate described score in predicting model of refinement;
Utilize data and the optimization method of database, from the parameter of score in predicting model described in the angle exercise of machine learning, to obtain the score in predicting parameters of formula of traffic behavior to timing scheme;
The described score in predicting parameters of formula that utilization obtains, in conjunction with described score in predicting model, the scoring of prediction traffic behavior to original timing scheme;
The historical used timing scheme of the described best timing scheme that contrast obtains and this traffic behavior, draws final optimal timing scheme.
2. the traffic signal optimization method based on hidden semantic model according to claim 1, the data of wherein said database are to gather by technology, or obtain by Traffic Administration Bureau; And the timing scheme in described database is by collecting, or generated by classic algorithm.
3. the traffic signal optimization method based on hidden semantic model according to claim 1, wherein, as the traffic indicators of scoring, chooses time delay as scoring.
4. the traffic signal optimization method based on hidden semantic model according to claim 1, the wherein said step for basic traffic behavior and the modeling of timing scheme comprises: traffic behavior and timing scheme are mapped to hidden semantic space with vectorial form, with mathematical linguistics, are expressed as:
q
t∈R
f，p
s∈R
f，
Wherein, vectorial q
trepresent timing scheme, vectorial p
srepresent traffic behavior, f represents interactional uncertain factor between traffic behavior and timing scheme.
5. the traffic signal optimization method based on hidden semantic model according to claim 1, the step of the accurate described score in predicting model of wherein said further refinement comprises:
The biasing scoring existing by equation expression traffic behavior and timing scheme is:
b
st＝u+b
s+b
t，
Thus, final score in predicting model is
Wherein,
for prediction appraisal result, b
strepresent biasing scoring, u represents average score, b
sand b
trepresent respectively traffic behavior and timing scheme marking itself or departed from the degree that draw is divided u, vectorial q by marking
trepresent timing scheme, vectorial p
srepresent traffic behavior.
6. the traffic signal optimization method based on hidden semantic model according to claim 1, wherein said data and the optimization method that utilizes database, from the parameter of score in predicting model described in the angle exercise of machine learning, to obtain traffic behavior, the step of the score in predicting parameters of formula of timing scheme is further comprised:
Arrangement extract traffic behavior, timing scheme and corresponding scoring in database all three-number set;
Data pre-service: the data of above-mentioned three-number set are processed into unified standard form on request;
The data of handling well are divided into M part, and wherein M-1 part, as training set, remains 1 part as test set;
Training pattern parameter on training set, selecting to minimize prediction scoring and true absolute mean square deviation of marking is herein target training,
Wherein, r
stfor true scoring, α represents the set of the tlv triple that traffic behavior on training set, timing scheme and corresponding scoring all exist, and in order to prevent the regularization term of over-fitting, λ is regularization parameter to second of above formula, and its occurrence is general to be chosen in conjunction with experiment; By random gradient descent method or least square method, be optimized herein, obtain respectively parameter to be asked;
Obtain after model, on test set, test the absolute mean square deviation under this model, by following formula, calculate:
Wherein, | α | represent the length of the set of the tlv triple that on test set, traffic behavior, timing scheme and corresponding scoring all exist;
Adjust parameter f, repeat above-mentioned two steps, train various model, and the absolute mean square deviation on corresponding test set;
According to above result, select to make the optimum prediction model of absolute mean square deviation RMSE minimum.
7. the traffic signal optimization method based on hidden semantic model according to claim 6, the step of the described best timing scheme that wherein said contrast obtains and the historical used timing scheme of this traffic behavior comprises: the described optimum prediction model that utilizes gained, for current traffic behavior is predicted the scoring of its original timing scheme, and the scoring contrast with the used timing scheme of tool by this scoring, determine final best timing scheme.
8. the traffic signal optimization method based on hidden semantic model according to claim 1, after being also included in and obtaining best timing scheme, further gives described database by the data feedback after practical application, further abundant or upgrade the step of described database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410360312.3A CN104112366B (en) | 2014-07-25 | 2014-07-25 | Method for traffic signal optimization based on latent semantic model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410360312.3A CN104112366B (en) | 2014-07-25 | 2014-07-25 | Method for traffic signal optimization based on latent semantic model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104112366A true CN104112366A (en) | 2014-10-22 |
| CN104112366B CN104112366B (en) | 2017-02-22 |
Family
ID=51709140
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410360312.3A Active CN104112366B (en) | 2014-07-25 | 2014-07-25 | Method for traffic signal optimization based on latent semantic model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104112366B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104794367A (en) * | 2015-05-12 | 2015-07-22 | 宁波克诺普信息科技有限公司 | Hospitalizing resource scoring and recommending method based on latent factor model |
| CN104965896A (en) * | 2015-06-23 | 2015-10-07 | 中国船舶重工集团公司第七0九研究所 | Recommendation method and system for representing user preference based on quadratic function |
| CN106297323A (en) * | 2016-09-30 | 2017-01-04 | 北京中星微电子有限公司 | Vehicle flow monitoring method and system |
| WO2020093702A1 (en) * | 2018-11-07 | 2020-05-14 | 南京邮电大学 | Deep q-network learning-based traffic light dynamic timing algorithm |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112037539B (en) * | 2020-07-31 | 2021-11-16 | 银江技术股份有限公司 | Method and system for recommending signal control scheme for saturated urban traffic network |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101281685A (en) * | 2008-01-30 | 2008-10-08 | 吉林大学 | Coordination control method for area mixed traffic self-adaption signal |
| EP2161698A1 (en) * | 2008-09-03 | 2010-03-10 | Siemens AG | Method for coordinating light signal-controlled nodes in a street network |
| CN101789178A (en) * | 2009-01-22 | 2010-07-28 | 中国科学院自动化研究所 | Optimized control method for traffic signals at road junction |
| CN103473291A (en) * | 2013-09-02 | 2013-12-25 | 中国科学院软件研究所 | Personalized service recommendation system and method based on latent semantic probability models |
| CN103903455A (en) * | 2014-04-14 | 2014-07-02 | 东南大学 | Urban road traffic signal control optimization system |
-
2014
- 2014-07-25 CN CN201410360312.3A patent/CN104112366B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101281685A (en) * | 2008-01-30 | 2008-10-08 | 吉林大学 | Coordination control method for area mixed traffic self-adaption signal |
| EP2161698A1 (en) * | 2008-09-03 | 2010-03-10 | Siemens AG | Method for coordinating light signal-controlled nodes in a street network |
| CN101789178A (en) * | 2009-01-22 | 2010-07-28 | 中国科学院自动化研究所 | Optimized control method for traffic signals at road junction |
| CN103473291A (en) * | 2013-09-02 | 2013-12-25 | 中国科学院软件研究所 | Personalized service recommendation system and method based on latent semantic probability models |
| CN103903455A (en) * | 2014-04-14 | 2014-07-02 | 东南大学 | Urban road traffic signal control optimization system |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104794367A (en) * | 2015-05-12 | 2015-07-22 | 宁波克诺普信息科技有限公司 | Hospitalizing resource scoring and recommending method based on latent factor model |
| CN104794367B (en) * | 2015-05-12 | 2018-01-12 | 宁波克诺普信息科技有限公司 | Medical treatment resource scoring based on hidden semantic model is with recommending method |
| CN104965896A (en) * | 2015-06-23 | 2015-10-07 | 中国船舶重工集团公司第七0九研究所 | Recommendation method and system for representing user preference based on quadratic function |
| CN104965896B (en) * | 2015-06-23 | 2018-10-02 | 中国船舶重工集团公司第七0九研究所 | The recommendation method and system of user preference are indicated based on quadratic function |
| CN106297323A (en) * | 2016-09-30 | 2017-01-04 | 北京中星微电子有限公司 | Vehicle flow monitoring method and system |
| WO2020093702A1 (en) * | 2018-11-07 | 2020-05-14 | 南京邮电大学 | Deep q-network learning-based traffic light dynamic timing algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104112366B (en) | 2017-02-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yu et al. | Prediction of highway tunnel pavement performance based on digital twin and multiple time series stacking | |
| Liu et al. | Contextual fuzzy cognitive map for decision support in geographic information systems | |
| CN109977237A (en) | A kind of dynamic law occurrence diagram spectrum construction method towards legal field | |
| CN104112366A (en) | Method for traffic signal optimization based on latent semantic model | |
| CN103678431A (en) | Recommendation method based on standard labels and item grades | |
| CN109919347A (en) | Road conditions generation method, relevant apparatus and equipment | |
| CN103455576B (en) | A kind of Network Learning Resource recommend method based on thinking map | |
| CN102541920A (en) | Method and device for improving accuracy degree by collaborative filtering jointly based on user and item | |
| CN104318340A (en) | Information visualization method and intelligent visual analysis system based on text curriculum vitae information | |
| CN109242149A (en) | A kind of student performance early warning method and system excavated based on educational data | |
| Lin et al. | A spatial-temporal hybrid model for short-term traffic prediction | |
| CN103116657A (en) | Individual searching method of network teaching resource | |
| CN105447145A (en) | Item-based transfer learning recommendation method and recommendation apparatus thereof | |
| Keyarsalan et al. | Designing an intelligent ontological system for traffic light control in isolated intersections | |
| CN104123847A (en) | ACP (artificial societies, computational experiments and parallel execution) method-based traffic signal recommendation system and corresponding method | |
| CN107545301A (en) | Page display method and device | |
| CN106997488A (en) | A kind of action knowledge extraction method of combination markov decision process | |
| Yang et al. | Feature selection in conditional random fields for map matching of GPS trajectories | |
| AlHalawani et al. | What makes London work like London? | |
| Cosido et al. | Automatic calculation of bicycle routes by combining meta-heuristics and GIS techniques within the framework of smart cities | |
| CN108694232A (en) | A kind of socialization recommendation method based on trusting relationship feature learning | |
| Axenie et al. | Fuzzy modelling and inference for physics-aware road vehicle driver behaviour model calibration | |
| CN110362828A (en) | Network information Risk Identification Method and system | |
| CN104657429A (en) | Complex-network-based technology-driven product innovation method | |
| Liu et al. | An adaptive traffic flow prediction model based on spatiotemporal graph neural network |
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 |