The Application Of Data Mining Techniques In Transport Sector

Since the reform and opening up, China’s economic have has a rapid development. The number of private cars gets a sharp increase. The car ownership of China have reached as much as104million so far. Then all kinds of traffic problems appeared. Frequent traffic accidents posed a serious threat to people’s property and lives. It’s inconvenience and wasting time if traffic jam is severe. Besides, it will reduce the well-being of everyone to some extent. Traditional methods, such as widen the road, increased traffic control personnel, has failed to meet the needs of the present stage. The rise of data mining technology in recent years has give us a hope to solve this problem.Progress and development of computer science and usage of modern hardware technologies in sampling data causes generating a huge amount of data to gather and process in most of the fields. During the last decade traffic management became a new field of science which produced unlimited data. This paper describes using the association rule mining to analyze the real cause of traffic accidents and using decision tree classification method to classify the collection data. They can provide decision support for traffic managers in the future and alleviate the traffic problems.This paper used the traffic data of Xinxiang City from December2011to December2012as data source. The collected data contains so many attributes that we use multi-level association rule mining method to excavate. So the traffic data is divided into three layers in which the attribute is thinner layer by layer. Then set the appropriate minimum support threshold to excavate layer by layer. And it will produce a collection of association rules. This can provide effective guidance for traffic managers so that road traffic accidents can be prevented to some extent. This paper also proposed using decision tree classification to the transport sector. It means to classify the collected data and find a lot of potential law from these large and disorder data. In order to illustrate the problem in a simple approach, so we adopt a simple example that is creating and interpret a decision tree. This example illustrate the application of the decision tree to the transport sector is feasible. And it can help traffic managers to make decisions, predict traffic new events and alleviate traffic jam.