粗集预处理数据的神经网络交通事件自动检测算法研究
|
摘要
随着高速公路的发展和大城市交通问题的日益严重,建立交通应急系统的要求也越来越迫切,而有效的交通事件检测是应急系统成功运行的第一步,也是关键之所在,所以对交通事件检测算法的研究就成为热点问题。本文试图把粗集理论和神经网络理论相结合,建立一种全新的粗集预处理数据的神经网络交通事件自动检测算法,主要研究内容包括以下几个方面:
总结现有的交通事件自动检测算法,对其性能进行比较分析,并详细地列举出评价算法优劣的评价指标和评价准则,提出本文欲建立算法的基本原理。
对粗集理论进行数据挖掘的若干问题进行了比较深入地研究,如属性特征的选择、连续属性离散化、和约简算法寻优等。对神经网络检测交通事件的几种算法也做了介绍和比较,详细研究了网络结构的选择和训练问题,为建立本文的算法提供了基础。
提出了建立粗集预处理数据的神经网络交通事件自动检测算法,该算法充分发挥了粗集理论和神经网络理论各自的优势。神经网络具有并行处理、网络全局、信息分布存储等特点,可通过训练、学习产生一个非线性映射,自适应地对数据进行聚类,同时具有较好的抑制噪声干扰的能力和较强的鲁棒性。其缺点是当输入信息空间的维数较大时,网络不仅结构复杂,而且训练时间也很长。而粗集可对数据进行属性约简和值约简,消除样本中的噪声和冗余对象。这两者的结合不仅可以减小网络的规模,同时通过消除对象冗余可减少网络的训练和学习负担,还可以通过消除噪声提高神经网络预测的准确性。本文算法中的神经网络采用的是多层前向反馈神经网络,并引入了动量项,采用自适应的学习率,可以获得更好的性能。
为检验算法的有效性,建立了事件仿真模型,获得所需数据,并用获得的交通流数据来训练和检测本文算法,对已有神经网络算法和传统检测算法与本文算法做了对比试验和分析,分析结果表明,本文算法可获得较高的检测率和较低的误报率,对评价指标的协调性也比其它算法更好。
With the development of highway and traffic problem being serious day by day inbig cities, the demand of establishing Emergence Management Systems (EMS)becomes urgent. Effective detection for traffic incident is the first step and the keycomponent of operating the EMS successfully;therefore, the study on detectionalgorithm for traffic incident has become a hot issue. The study tries to combine therough set theory and neural network to establish a new automatic detection algorithmfor traffic incident using neural network based on rough set filtration. Main researchwork is listed as follows:
The existing automatic detection algorithms for traffic incident are summarized,the performances of it are compared and analyzed, the index for evaluatingalgorithm's performance is listed,and then the basic principle of the algorithm in thethesis is expounded.
A number of problems in data mining using rough set had been studied deeply inthe thesis,for example, feature selection, consecutive feature discretization, seekingthe optimum deduction algorithm. Several algorithms using neural network to detecttraffic incident also had been introduced and compared. The structure's selection andtraining of neural network were studied in particular, which were the basic ofestablishing the algorithm in the thesis.
A new automatic detection algorithm for traffic incident using neural networkbased on rough set filtration was proposed, which has the advantages of neuralnetwork and rough set theory. Neural network has several characteristics, such asdisposing datum in parallel, aiming for overall function, storing informationdistributed and so on, which can produce a non-linear map by training and learning,cluster data adaptively, with the abilities of restraining the noise's disturbance andgood robustness. Its shortcomings are that when the space dimension of the inputinformation is larger, not only complex the network's structure is, but also the networkneed more time to train. Rough set can deduct feature and value of the data,eliminatethe noise and redundant targets in sample. The merge not only reduced the scale of thenetwork, lessened burden of training and learning by eliminating redundant targets,but also improved the accuracy of detection by eliminating noise. The network in thethesis has multi-layer feed forward architecture;the motion vector and the adaptive
learning rate are also adopted for better performance.In order to verify the effect of the algorithm proposed in the thesis, simulationmodel of traffic incident was established to obtain the traffic data which was used totrain and test the algorithm. Comparing and analyzing had been done among thealgorithm in the thesis, the existing neural network and traditional detectionalgorithms. The results show that the algorithm in the thesis has higher detection rateand lower false detection rate. Its coordination ability to evaluation index is alsobetter than other algorithms.
总结现有的交通事件自动检测算法,对其性能进行比较分析,并详细地列举出评价算法优劣的评价指标和评价准则,提出本文欲建立算法的基本原理。
对粗集理论进行数据挖掘的若干问题进行了比较深入地研究,如属性特征的选择、连续属性离散化、和约简算法寻优等。对神经网络检测交通事件的几种算法也做了介绍和比较,详细研究了网络结构的选择和训练问题,为建立本文的算法提供了基础。
提出了建立粗集预处理数据的神经网络交通事件自动检测算法,该算法充分发挥了粗集理论和神经网络理论各自的优势。神经网络具有并行处理、网络全局、信息分布存储等特点,可通过训练、学习产生一个非线性映射,自适应地对数据进行聚类,同时具有较好的抑制噪声干扰的能力和较强的鲁棒性。其缺点是当输入信息空间的维数较大时,网络不仅结构复杂,而且训练时间也很长。而粗集可对数据进行属性约简和值约简,消除样本中的噪声和冗余对象。这两者的结合不仅可以减小网络的规模,同时通过消除对象冗余可减少网络的训练和学习负担,还可以通过消除噪声提高神经网络预测的准确性。本文算法中的神经网络采用的是多层前向反馈神经网络,并引入了动量项,采用自适应的学习率,可以获得更好的性能。
为检验算法的有效性,建立了事件仿真模型,获得所需数据,并用获得的交通流数据来训练和检测本文算法,对已有神经网络算法和传统检测算法与本文算法做了对比试验和分析,分析结果表明,本文算法可获得较高的检测率和较低的误报率,对评价指标的协调性也比其它算法更好。
With the development of highway and traffic problem being serious day by day inbig cities, the demand of establishing Emergence Management Systems (EMS)becomes urgent. Effective detection for traffic incident is the first step and the keycomponent of operating the EMS successfully;therefore, the study on detectionalgorithm for traffic incident has become a hot issue. The study tries to combine therough set theory and neural network to establish a new automatic detection algorithmfor traffic incident using neural network based on rough set filtration. Main researchwork is listed as follows:
The existing automatic detection algorithms for traffic incident are summarized,the performances of it are compared and analyzed, the index for evaluatingalgorithm's performance is listed,and then the basic principle of the algorithm in thethesis is expounded.
A number of problems in data mining using rough set had been studied deeply inthe thesis,for example, feature selection, consecutive feature discretization, seekingthe optimum deduction algorithm. Several algorithms using neural network to detecttraffic incident also had been introduced and compared. The structure's selection andtraining of neural network were studied in particular, which were the basic ofestablishing the algorithm in the thesis.
A new automatic detection algorithm for traffic incident using neural networkbased on rough set filtration was proposed, which has the advantages of neuralnetwork and rough set theory. Neural network has several characteristics, such asdisposing datum in parallel, aiming for overall function, storing informationdistributed and so on, which can produce a non-linear map by training and learning,cluster data adaptively, with the abilities of restraining the noise's disturbance andgood robustness. Its shortcomings are that when the space dimension of the inputinformation is larger, not only complex the network's structure is, but also the networkneed more time to train. Rough set can deduct feature and value of the data,eliminatethe noise and redundant targets in sample. The merge not only reduced the scale of thenetwork, lessened burden of training and learning by eliminating redundant targets,but also improved the accuracy of detection by eliminating noise. The network in thethesis has multi-layer feed forward architecture;the motion vector and the adaptive
learning rate are also adopted for better performance.In order to verify the effect of the algorithm proposed in the thesis, simulationmodel of traffic incident was established to obtain the traffic data which was used totrain and test the algorithm. Comparing and analyzing had been done among thealgorithm in the thesis, the existing neural network and traditional detectionalgorithms. The results show that the algorithm in the thesis has higher detection rateand lower false detection rate. Its coordination ability to evaluation index is alsobetter than other algorithms.
引文
[1] 杨兆升,智能运输系统概论[M],北京:人民交通出版社,2003,130~132.
[2] 姜紫峰,刘小坤,基于神经网络的交通事件检测方法[J],西安公路交通大学学报,2000,20(3):67~70.
[3] 贺国光,ITS 系统工程导论[M], 北京:中国铁道出版社,2004,53~85.
[4]Karl F, Petty et al, A New Methodology for Evaluating Incident Detection Algorithms[J],Transportation Research,2002,22(2):189~204.
[5]Baher Abdulhai, Stephen G. Ritchie, Enhancing the Universality and Transferability of Freeway Incident Detection Using a Bayesian-based Neural Network, Transportation Research Part C,1999,C(7):261~280.
[6] 陆光辉,肖人彬,基于粗集理论和神经网络的集成化数据挖掘方法研究[J],小型微型计算机系统,2002,23(5):552~557.
[7]曾黄麟,粗集理论集及应用[M],重庆:重庆大学出版社,1996,15~60.
[8]Z. Pawlak, Rough Sets, Theoretical Aspects of Reasoning about Data[M], Dordrecht: Kluwer Academic Publishers,1991,23~79.
[9]S.Tsumoto and H.Tanaka, AQ, Rough Sets, and Matroid Theory, Proc.of RSKD'93, Japan, 1993,34~62.
[10]S.K.M wong et al, Rough Sets: Probabilistic versus Deterministic Approach, Int J. Man-Machine studies,1998, 13(3):245~262.
[11]A. Skowron, The Rough Sets Theory and Evidence Theory, Fundamental Informaticae, 1990, 13(2):245~262.
[12]Pawlak Z, Grzymala-Busse J, Slowinski R etc, Rough sets, Communiciations on the ACM,1995,38(11): 89~95.
[13]李仁璞,王正欧,基于粗集理论和神经网路结合的数据挖掘新方法[J],情报学报,2002, 21,(6): 674~679.
[14]Beynon M, Reducts within the variable precision rough sets model, A further investigation[J],European Journal of Operational Research,2001, 134(3): 592~605.
[15]Quafafou M, α -RST, A Generalization of Rough set theory[J], Information Scinences,2000,124(1-4): 301~316.
[16]Sironmoney A, Inoue K, Consistency and Completeness in Rough Sets[J], Journal of Information Systems,2000,15(3): 207~220.
[17]Guan J W, Bell D A, Rough Computational Methods for Information Systems [J], Artificial Intelligence, 1998, 105(1-2): 77~103.
[18]刘业政,杨善林,马溪郡,粗糙集数据分析的计算方法[J],合肥工业大学学报,2002,25(2):161~166.
[19]Starzyk J A, Nelson D E, Sturtz K,A Mathematical Foundation for Improved Reduct Generation in Information Systems[J],Knowledge and Information Systems,2000, 2(2): 131~146.
[20]Pawlak Z, Rough sets, Decision Algorithms and Bayes' Theorem[J], European Journal of Operational Research,2002,136(1): 181~189.
[21]Golan R H, Ziarko W, Methodology for Stock Market Analysis Utilizing Rough Set Theory, Proceedings of the IEEE/IAEE 1995 Computational Intelligence for Financial Engineering, New York: IEEE, 1995, 32~40.
[22]Yao Y Y. Generalized Rough Set Models, L Polkowski, A Skowron Eds, Rough Sets in Knowledge Discovery 1-Methodology and Applications, Physica-Verlag Heidelberg, New York:1998,287~318.
[23]Pawlak Z. Rough set, Theoretical Aspects of Reasoning about Data, Boston, London, Dordrecht: Kluwer Publishers, 1991, 69~90.
[24]Pawlak Z. Rough sets, Rough Function and Rough Calculus, S K Pal, A Skowron eds, Rough-fuzzy Hybridization, a New Trend in Decision-making, Springer-Verlag, Singapore:1999,99~109.
[25]Ziarko W. Variable Precision Rough Sets Mode [J], Journal of Computer System Science, 1993,25(4):39~50.
[26]张文修,吴伟志等,粗糙集理论与方法[M],北京:科学出版社,2001,132~157.
[27]Polkowski L,A Skowron, Rough Mereology, Proceedings of the Symposium on Methodologies for Intelligent Systems, Charlotte, NC, Springer-Verlag, Berlin:1994,85~94.
[28]王钰,苗夺谦,周育健,关于 Rough Set 理论与应用的综述[J],模式识别与人工智能,1996,9(4):337~344.
[29]Pawlai Z, Grzymala-Busse J, Slowinski R ect, Rough Sets[J], Communications on the ACM, 1995,38(11):89~95.
[30]Pawlak Z, Rough sets, Theoretical Aspects of Reasoning about Data, Dordrecht: Kluwer Academic Publishers, 1991, 67~98.
[31]Selwyn Piramuthu, Evaluating Feature Selection Methods for Learning in Data Mining Applications[J], European Journal of Operational Research , 2004,156(8): 483~494.
[32] Ron Kohavi, Gerorge H John, Wrappers for feature subset selection [J], Artificial Intelligence, 1997,97(7): 273~324.
[33]李仁璞,分类数据挖掘中若干基本问题的研究:[博士学位论文],天津:天津大学,2003,8~17.
[34]Holte R C.,Very Simple Classification Rules Perform well on Most Commonly Used Datasets [J], Machine learning,1993,11(5):63~90.
[35]John G H, Kohavi R, Pfloger K., Irrelevant Features and the Subset Selection Problem ,Mitchell T M. Proceedings on Machine Learning'94, Morgan Koffmann Publishers, 1994,121~129.
[36]潘韬,赵卫东,盛昭瀚,决策表最优属性子集的选择[J],东南大学学报,2000,31(5):118~122.
[37]陈彬,洪家荣,王亚东,最优特征矩阵选择问题[J],计算机学报,1997,20(2):133~138.
[38]赵卫东,戴伟辉,基于特征矩阵的决策表约简研究[J],系统工程理论与实践,2003,3(3):65~69.
[39]王亚英,张春慨,绍惠鹤,启发式知识约简算法的研究与应用[J],控制与决策,2001,16(6):886~889.
[40]朱红,基于 Rough Set 的属性及属性值约简的一种算法[J],湘潭大学自然科学学报,2002,24(3):36~39.
[41]Hu XiaoHua, Knowledge Discovery in Databases, An Attribute-oriented Rough Set Approach, Regina University, 1995,128~147.
[42]王钰,王任,苗多谦等,基于 Rough Set 理论的属性约简及规则提取方法[J],软件学报,1999,10(11):1206~1211.
[43]Wang J, Wang J, Reduction Algorithms Based on Discernability Matrix, The Ordered Attributes Method[J], Journal of Computer Science and Technolgy,2001, 16(6): 489~504.
[44]Phuong N H, Phong L L, Santiprahhob P, et al, Approach to Generating Rules for Expert Systems Using Rough Set Theory, Proceedings of the Ninth International Conference on Fuzzy Systems Association Congress, Vancouver: IEEE Press, 2001,877~882.
[45]吕琪,王慧,基于动态神经网络模型的交通事件检测算法[J],公路交通科技,2003, 20(6): 105~108.
[46]Ko Ito, Applying Neural Network to Automatic Incident Detection[J], Traffic Technology International,1996, 22(3):85~102.
[47]Stephen G. Ritchrie et al, Simulation of Freeway Incident Detection using Artificial Neural Networks[J],Transportation Research,1993,19(2): 203~217.
[48]Edmond Chin-ping Chang et al, Incident Detection Using Advanced Technologies, Presentation at the 72nd Annual TRB Meeting,1993, 25~43.
[49]Ardeshir Faghri et al, Evaluation of Artificial Neural Network Applications in Transportation Engineering, Transportation Research Record 1358,1992,51~73.
[50]Cheu. R.L. et al, Investigations of a neural network model for freeway incident detection[J], Artificial Intelligence,1991,12(4):267~264.
[51]周伟,罗石贵,基于模糊综合识别的事件检测算法[J],西安公路交通大学学报,2001,21(2):70~73.
[52]周雪铭,基于小波分析和神经网络的交通事件自动检测[J],公路与汽运,2002,91(4):17~19.
[53]Cheu. R. L. et al, Automated Detection of Lane-blocking Freeway Incidents Using Artificial Neural Networks [J], Transportation Research C, 1995, 371~388.
[54]Kohonen T, Self-organized Formation of Topologically Correct Feature maps, Biol Cybernet, 1982, 43:59~69.
[55]Ritchie H , Martinetz T. and Schulten K., Neural Computation and Self-organizing Maps, Addison-Wesley, Reading,MA, 1992, 17~50.
[56]Grossberg S, Adaptive Pattern Classification and Universal Recording, I, Parallel Development and Coding of Neural Feature Detectors, Biol. Cybernet, 1976,23: 121~134.
[57]Grossberg S.,Adaptive Pattern Classification and Universal Recording II,Feedback,Expectation,Olfaction,Illusions,Biol,Cybernet,1976,187~202.
[58]Capenter G. A. and Grossberg S., ART2, Self-organization of Stable Category Recognition Codes for Analog Input Patterns, 1987, 26:4919~4930.
[59] Xin Jin et al, Development and Adaptation of Constructive Probabilistic Neural Network In Freeway Incident Detection[J] ,Transportation Research,2002,34(4): 121~147.
[60]Payne H J.FREFLO, A Macroscopic Simulation Model for Freeway Traffic [J]. Transportation Research C1, 1993(3):203~217.
[61] Cybenko G,Approximation by Superposition of Sigmoidal Functions[J],Mathematics of Control Signals and Systems,1989,2(4): 303~314.
[62]Yasui S. Convergence suppression and Divergence Facilitation, Minimum and Joint Use of Hidden Units by Multiple Outputs [J], Neural Networks, 1997, 10(2):353~367.
[63]Anders U.Korn O, Model Selection in Neural Networks [J],Neural Networks,1999,12(2):309~323.
[64] 辛旭明,孙福良,粗集在知识发现中的应用[J],东华大学学报(自然科学版),2003,29(5):131~134.
[65]胡丹,莫智文,粗糙集理论在模糊网络中的应用研究[J],四川师范大学学报(自然科学版),2002,25(1):36~38.
[66]Lingras P J, Rough neural networks, Proceedings of Sixth International Conference on Information, Processing and Management of Uncertainty in Knowledge-Based Systems, Canada, Spain:1996,1445~1450.
[67]杨慧中,陶振麟,带一类自适应非线性的变步长 BP 学习算法[J],系统仿真学报,2001,8(13):108~110.
[2] 姜紫峰,刘小坤,基于神经网络的交通事件检测方法[J],西安公路交通大学学报,2000,20(3):67~70.
[3] 贺国光,ITS 系统工程导论[M], 北京:中国铁道出版社,2004,53~85.
[4]Karl F, Petty et al, A New Methodology for Evaluating Incident Detection Algorithms[J],Transportation Research,2002,22(2):189~204.
[5]Baher Abdulhai, Stephen G. Ritchie, Enhancing the Universality and Transferability of Freeway Incident Detection Using a Bayesian-based Neural Network, Transportation Research Part C,1999,C(7):261~280.
[6] 陆光辉,肖人彬,基于粗集理论和神经网络的集成化数据挖掘方法研究[J],小型微型计算机系统,2002,23(5):552~557.
[7]曾黄麟,粗集理论集及应用[M],重庆:重庆大学出版社,1996,15~60.
[8]Z. Pawlak, Rough Sets, Theoretical Aspects of Reasoning about Data[M], Dordrecht: Kluwer Academic Publishers,1991,23~79.
[9]S.Tsumoto and H.Tanaka, AQ, Rough Sets, and Matroid Theory, Proc.of RSKD'93, Japan, 1993,34~62.
[10]S.K.M wong et al, Rough Sets: Probabilistic versus Deterministic Approach, Int J. Man-Machine studies,1998, 13(3):245~262.
[11]A. Skowron, The Rough Sets Theory and Evidence Theory, Fundamental Informaticae, 1990, 13(2):245~262.
[12]Pawlak Z, Grzymala-Busse J, Slowinski R etc, Rough sets, Communiciations on the ACM,1995,38(11): 89~95.
[13]李仁璞,王正欧,基于粗集理论和神经网路结合的数据挖掘新方法[J],情报学报,2002, 21,(6): 674~679.
[14]Beynon M, Reducts within the variable precision rough sets model, A further investigation[J],European Journal of Operational Research,2001, 134(3): 592~605.
[15]Quafafou M, α -RST, A Generalization of Rough set theory[J], Information Scinences,2000,124(1-4): 301~316.
[16]Sironmoney A, Inoue K, Consistency and Completeness in Rough Sets[J], Journal of Information Systems,2000,15(3): 207~220.
[17]Guan J W, Bell D A, Rough Computational Methods for Information Systems [J], Artificial Intelligence, 1998, 105(1-2): 77~103.
[18]刘业政,杨善林,马溪郡,粗糙集数据分析的计算方法[J],合肥工业大学学报,2002,25(2):161~166.
[19]Starzyk J A, Nelson D E, Sturtz K,A Mathematical Foundation for Improved Reduct Generation in Information Systems[J],Knowledge and Information Systems,2000, 2(2): 131~146.
[20]Pawlak Z, Rough sets, Decision Algorithms and Bayes' Theorem[J], European Journal of Operational Research,2002,136(1): 181~189.
[21]Golan R H, Ziarko W, Methodology for Stock Market Analysis Utilizing Rough Set Theory, Proceedings of the IEEE/IAEE 1995 Computational Intelligence for Financial Engineering, New York: IEEE, 1995, 32~40.
[22]Yao Y Y. Generalized Rough Set Models, L Polkowski, A Skowron Eds, Rough Sets in Knowledge Discovery 1-Methodology and Applications, Physica-Verlag Heidelberg, New York:1998,287~318.
[23]Pawlak Z. Rough set, Theoretical Aspects of Reasoning about Data, Boston, London, Dordrecht: Kluwer Publishers, 1991, 69~90.
[24]Pawlak Z. Rough sets, Rough Function and Rough Calculus, S K Pal, A Skowron eds, Rough-fuzzy Hybridization, a New Trend in Decision-making, Springer-Verlag, Singapore:1999,99~109.
[25]Ziarko W. Variable Precision Rough Sets Mode [J], Journal of Computer System Science, 1993,25(4):39~50.
[26]张文修,吴伟志等,粗糙集理论与方法[M],北京:科学出版社,2001,132~157.
[27]Polkowski L,A Skowron, Rough Mereology, Proceedings of the Symposium on Methodologies for Intelligent Systems, Charlotte, NC, Springer-Verlag, Berlin:1994,85~94.
[28]王钰,苗夺谦,周育健,关于 Rough Set 理论与应用的综述[J],模式识别与人工智能,1996,9(4):337~344.
[29]Pawlai Z, Grzymala-Busse J, Slowinski R ect, Rough Sets[J], Communications on the ACM, 1995,38(11):89~95.
[30]Pawlak Z, Rough sets, Theoretical Aspects of Reasoning about Data, Dordrecht: Kluwer Academic Publishers, 1991, 67~98.
[31]Selwyn Piramuthu, Evaluating Feature Selection Methods for Learning in Data Mining Applications[J], European Journal of Operational Research , 2004,156(8): 483~494.
[32] Ron Kohavi, Gerorge H John, Wrappers for feature subset selection [J], Artificial Intelligence, 1997,97(7): 273~324.
[33]李仁璞,分类数据挖掘中若干基本问题的研究:[博士学位论文],天津:天津大学,2003,8~17.
[34]Holte R C.,Very Simple Classification Rules Perform well on Most Commonly Used Datasets [J], Machine learning,1993,11(5):63~90.
[35]John G H, Kohavi R, Pfloger K., Irrelevant Features and the Subset Selection Problem ,Mitchell T M. Proceedings on Machine Learning'94, Morgan Koffmann Publishers, 1994,121~129.
[36]潘韬,赵卫东,盛昭瀚,决策表最优属性子集的选择[J],东南大学学报,2000,31(5):118~122.
[37]陈彬,洪家荣,王亚东,最优特征矩阵选择问题[J],计算机学报,1997,20(2):133~138.
[38]赵卫东,戴伟辉,基于特征矩阵的决策表约简研究[J],系统工程理论与实践,2003,3(3):65~69.
[39]王亚英,张春慨,绍惠鹤,启发式知识约简算法的研究与应用[J],控制与决策,2001,16(6):886~889.
[40]朱红,基于 Rough Set 的属性及属性值约简的一种算法[J],湘潭大学自然科学学报,2002,24(3):36~39.
[41]Hu XiaoHua, Knowledge Discovery in Databases, An Attribute-oriented Rough Set Approach, Regina University, 1995,128~147.
[42]王钰,王任,苗多谦等,基于 Rough Set 理论的属性约简及规则提取方法[J],软件学报,1999,10(11):1206~1211.
[43]Wang J, Wang J, Reduction Algorithms Based on Discernability Matrix, The Ordered Attributes Method[J], Journal of Computer Science and Technolgy,2001, 16(6): 489~504.
[44]Phuong N H, Phong L L, Santiprahhob P, et al, Approach to Generating Rules for Expert Systems Using Rough Set Theory, Proceedings of the Ninth International Conference on Fuzzy Systems Association Congress, Vancouver: IEEE Press, 2001,877~882.
[45]吕琪,王慧,基于动态神经网络模型的交通事件检测算法[J],公路交通科技,2003, 20(6): 105~108.
[46]Ko Ito, Applying Neural Network to Automatic Incident Detection[J], Traffic Technology International,1996, 22(3):85~102.
[47]Stephen G. Ritchrie et al, Simulation of Freeway Incident Detection using Artificial Neural Networks[J],Transportation Research,1993,19(2): 203~217.
[48]Edmond Chin-ping Chang et al, Incident Detection Using Advanced Technologies, Presentation at the 72nd Annual TRB Meeting,1993, 25~43.
[49]Ardeshir Faghri et al, Evaluation of Artificial Neural Network Applications in Transportation Engineering, Transportation Research Record 1358,1992,51~73.
[50]Cheu. R.L. et al, Investigations of a neural network model for freeway incident detection[J], Artificial Intelligence,1991,12(4):267~264.
[51]周伟,罗石贵,基于模糊综合识别的事件检测算法[J],西安公路交通大学学报,2001,21(2):70~73.
[52]周雪铭,基于小波分析和神经网络的交通事件自动检测[J],公路与汽运,2002,91(4):17~19.
[53]Cheu. R. L. et al, Automated Detection of Lane-blocking Freeway Incidents Using Artificial Neural Networks [J], Transportation Research C, 1995, 371~388.
[54]Kohonen T, Self-organized Formation of Topologically Correct Feature maps, Biol Cybernet, 1982, 43:59~69.
[55]Ritchie H , Martinetz T. and Schulten K., Neural Computation and Self-organizing Maps, Addison-Wesley, Reading,MA, 1992, 17~50.
[56]Grossberg S, Adaptive Pattern Classification and Universal Recording, I, Parallel Development and Coding of Neural Feature Detectors, Biol. Cybernet, 1976,23: 121~134.
[57]Grossberg S.,Adaptive Pattern Classification and Universal Recording II,Feedback,Expectation,Olfaction,Illusions,Biol,Cybernet,1976,187~202.
[58]Capenter G. A. and Grossberg S., ART2, Self-organization of Stable Category Recognition Codes for Analog Input Patterns, 1987, 26:4919~4930.
[59] Xin Jin et al, Development and Adaptation of Constructive Probabilistic Neural Network In Freeway Incident Detection[J] ,Transportation Research,2002,34(4): 121~147.
[60]Payne H J.FREFLO, A Macroscopic Simulation Model for Freeway Traffic [J]. Transportation Research C1, 1993(3):203~217.
[61] Cybenko G,Approximation by Superposition of Sigmoidal Functions[J],Mathematics of Control Signals and Systems,1989,2(4): 303~314.
[62]Yasui S. Convergence suppression and Divergence Facilitation, Minimum and Joint Use of Hidden Units by Multiple Outputs [J], Neural Networks, 1997, 10(2):353~367.
[63]Anders U.Korn O, Model Selection in Neural Networks [J],Neural Networks,1999,12(2):309~323.
[64] 辛旭明,孙福良,粗集在知识发现中的应用[J],东华大学学报(自然科学版),2003,29(5):131~134.
[65]胡丹,莫智文,粗糙集理论在模糊网络中的应用研究[J],四川师范大学学报(自然科学版),2002,25(1):36~38.
[66]Lingras P J, Rough neural networks, Proceedings of Sixth International Conference on Information, Processing and Management of Uncertainty in Knowledge-Based Systems, Canada, Spain:1996,1445~1450.
[67]杨慧中,陶振麟,带一类自适应非线性的变步长 BP 学习算法[J],系统仿真学报,2001,8(13):108~110.