In order to further improve the performances and stabilization of multiple kernel support machine (MKL-SVM) in traffic incident detection, this paper presents a new algorithm called MKL-SVM ensemble. The proposed algorithm uses the bagging technique to train different individual MKL-SVM classifiers, then takes the weighted voting way to combine the output of the individual MKL-SVM classifiers. Some experiments have been performed to evaluate the performances of the four algorithms: standard SVM, SVM ensemble, MKL-SVM and the proposed algorithm (MKL-SVM ensemble). The experimental results show that the proposed algorithm has the best comprehensive performances in traffic incident detection. More important, the performances of the proposed algorithm are very stable. Meanwhile, in order to achieve relatively better performances, the proposed algorithm need less individual classifiers to construct the ensemble than SVM ensemble algorithm. Thus, compared to SVM ensemble algorithm, the complexity of the ensemble classifier of the proposed algorithm is reduced greatly. Conveniently, the proposed algorithm also avoids the burden of selecting the appropriate kernel function and parameters.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Traffic incident detection by multiple kernel support vector machine ensemble


    Contributors:
    Xiao, Jianli (author) / Liu, Yuncai (author)


    Publication date :

    2012-09-01


    Size :

    189935 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Traffic Incident Detection Using Multiple-Kernel Support Vector Machine

    Xiao, Jianli / Liu, Yuncai | Transportation Research Record | 2012



    Support vector machine models for freeway incident detection

    Ruey Long Cheu, / Srinivasan, D. / Eng Tian Teh, | IEEE | 2003


    Support Vector Machine Models in Freeway Incident Detection

    Cheu, R. L. / Srinivasan, D. / Teh, E. T. et al. | British Library Conference Proceedings | 2003