Data Mining Technology In Traffic Management Applications

Prediction of Short-time traffic flow volume and evaluation of region traffic safety are important for traffic management. This dissertation mainly focuses on the above works using data mining techniques. The main work of the dissertation includes:Based on the study of data mining technology and Intelligent Traffic Management Platform, the Integrated Traffic Information Platform in Hangzhou City is proposed combined with the current traffic situation and long-range traffic planning of Hangzhou City. This Platform can obtain useful knowledge from the all kinds of traffic datasets.According to the Traffic Prediction Module of Integrated Traffic Information Platform in Hangzhou City, using support vector machine (SVM) with a time-dependent structure, a new model is proposed to predict short-time traffic flow volume. In order to match the time varying characteristic of the traffic flow volume, in the developed model, each prediction requires a reconstruction process of SVM structure. The current SVM structure is determined by restraining with the input of the data of the traffic flow volume in the last hour. Then the predicted value is obtained according to the current SVM structure. The experimental results show that the prediction model with a time-dependent structure SVM outperforms the one without a time-dependent structure. According to the Decision-making Module of Integrated Traffic Information Platform in Hangzhou City, a model is developed to evaluate the region traffic safety using Principal Component Analysis (PCA). Traffic safety index obtained by the proposed model is a good measure to the level of the region traffic safety. The main difficulty of this model is how to choose the evaluation data basis because of the fact that different data basis has different evaluated results. In this work, two kinds of typical datasets, including basic dataset and extended dataset, are investigated. The case study indicates that model based on the extended dataset has more consistent evaluated result with the practical traffic environment than the one based on the basic dataset.