Research On The Database And Data Mining Method For The Causes Of Traffic Accident

The road traffic safety is one of the essential factors to safeguard the social stability and promoting economic development. However, the road traffic accidents cast a shadow over the social progress and development. Therefore, no matter in the developed or the developing countries traffic accident is to reseach as a major social issues.This paper firstly has analysed the Chinese severe situation of the road traffic safety and believes that China has entered a road traffic accident-prone period. It is significantly meaningful to analyse the causes of traffic accidents, which can effectively prevent and reduce the times of the traffic accident,to provide the technical support for traffic management decision-making and to protect people's lives and property.Then, this paper ,which starts from the road traffic accident database and has analysed comparatively the database record of China , the United States, Japan, the European Union and other countries and regions ,believes that the Chinese road traffic accident database needs to be further improved and optimized in order to be more reasonable to explain the causes of road traffic accident.Road traffic accident data in the database provides convenience for the job of analysing the causes . However, because of the traffic accident data collection records and the existing issues when collecting, it makes the traffic accident database contain an error in the data, or obviously inconsistent logic, or manual inputting error and so on. Therefore, before data processing, it is an extremely important aspect to preprocess the database.The paper ,which is based on the existing number pre-processing technology and combines with the actual needs of the job of causes analysis, puts forward the data pre-processing technology, which adapts to the causes analysis of the road traffic accidents, that is, the method of analysis the data cleaning logic. The ideas of accident data pre-processing technology ensures the authenticity and reliability of the accident data to the maximum extent ,which lays a good foundation for the accidental cause analysis. We try to look for a more reasonable method in data mining method when we process the data,in order to unfold the traffic course law. So we have try many mining method include the principal component analysis,Rough Set Theory,neural network algorithm and Association Rules . from the result of Rough Set Theory and the Association Rules integrated database,and the result are stable.it can satisfy me for the need of analysis in the traffic field.In the end , this paper gives example about analysis of the road traffic accident data–base and mining , introducing the analysis method and steps. it is a positive taste for the great large number analysis by the data mining method.we also wish to obtain the advice and suggestion in the deep research .