01514328

Component

This study presents the applicability of Naïve Bayes classifier ensemble for traffic incident detection. The standard Naive Bayes (NB) has been applied on traffic incident detection and achieved good results. However, the detection result of the practically implemented NB depends on the choosing of the optimal threshold, which is determined mathematically by using Bayesian concepts in the incident-detection process. To avoid the burden of choosing the optimal threshold and tuning the parameters, and furthermore, to improve the limited classification performance of the NB and enhance the detection performance, the authors propose to apply the NB classifier ensemble to incident detection. The NB classifier ensemble algorithm trains many individual NB classifiers to construct the classifier ensemble, then uses this classifier ensemble to detect the traffic incidents. Consequently, it needs to train many times. In addition, the authors also propose to combine Naïve Bayes and decision tree (NBTree) to detect incidents. Different from NB, NBTree is a hybrid approach that attempts to utilize the advantage of both decision trees and naïve Bayesian classifier. Compared with NB ensemble algorithm, the training time cost of NBTree is much lower, which is because NBTree only needs to train one time. In this paper, extensive experiments have been performed to evaluate the performances of the three algorithms: standard NB, NB ensemble, NBTree. The experimental results show that the performances of five rules of NB classifier ensemble are significantly better than standard NB and slightly better than NBTree in some indicators. More important, the performances of NB classifier ensemble are very stable.

This paper was sponsored by TRB committee ABJ80 Statistical Methods.

01503729

14-1014

English

22p

2014

Transportation Research Board 93rd Annual Meeting

Digital/other

Figures; References; Tables

Algorithms; Bayes' theorem; Decision trees; Incident detection; Traffic incidents

Highways; Operations and Traffic Management; Planning and Forecasting; I72: Traffic and Transport Planning; I73: Traffic Control

Transportation Research Board Annual Meeting 2014 Paper #14-1014

TRIS, TRB, ATRI

Jan 27 2014 2:24PM