The Application Of The Mining Of Association Rules In Analysis Of Traffic Accidents

In the 21st century, with the increase of the auto possession, the traffic accidents which occur frequently have become a serious social problem. Of course, the cause of the traffic accident is various, but it is particularly important whether we can make a timely and scientific analysis of the traffic accident which will directly affect the prevention from happening the similar tragedy in future. With the comprehensive and high-speed economic development in the meantime, the traffic accident has showed diversity and complexity, which means that the data analysis, the data processing tools, the way and the method in the traditional sense have been completely unable to meet the needs of the times.Data mining can be defined as that the knowledge is extracted or dredged out from the huge unknown data and association rules is the most important methods of the data mining now. Association rules reflect the connections and mutual dependence between things, which denotes that data support and confidence is satisfied by the given value (threshold).Apriori arithmetic is very classic in mining association rule, which contains an important property that every loophole in the frequent item-sets shall be frequent". Through the professional knowledge and technology, association rules can be used to analyze the data and give the regulatory prediction so as to find the inner causal relationship. It is valuable that the data relationship can be found from the mass of data for the decision analysis.With the data mining technology, the useful information can be find from every detail of the data which records the traffic accidents, that is to say, combination of several factors intrinsically linked to the accident can be found in the rule may result in traffic accidents. If the inner link rules between the accidents are found, then Driving Behavior can be humanly controlled and intervened, some accident condition will disappear and the probability of traffic accident will be artificially loweringAs road traffic accidents triggers the complexity of the classic data mining Apriori algorithm is not suitable for non-single-dimensional rule mining, it needs to expand. This chapter with practical problems, the problems of our study was designed to simplify at the same time an improved Apriori algorithm, and support multi-dimensional association rule mining, the introduction of multi-dimensional attributes, so dig out the proposed rules would be more practical and of.In the multi-dimensional database mining of association rules as a predicate for each dimension, a more detailed description of the facts. Required information in the mining, the need to search for frequent massive database subset. In the multi-level multi-dimensional data model, a subset of the search frequency setting required for each layer of a size suitable minimum support, so the actual operation of the more complicated, it can be around a bend to streamline operations. Approach is to analyze the rules first before we choose one particular level for the correlation analysis and the exclusion of other levels of analysis, so this problem is reduced to a single rule analysis. For example, we want to tap the accident location, select the associated objects as county roads layer, the rest of the dimensions are so selected. This problem is simplified, and also of our multi-dimensional multi-level road safety accident data mining purposes.This paper was based on road traffic accidents which became a social problem. Single Apriori arithmetic which is more familiar to people is improved, the description of common attributes to traffic accident is given, the model of road traffic accidents organization was set up by the star model for the data connection. The large number of the data source of traffic accident was organized for the attribute information mining, which is useful for digging out the various complex relationships. Through the data mining of association rules technology for analyzing data, some traffic accident rules was found. The authors described the stress of the definition of the vehicle accident and the algorithm for the extraction of the data association rulesIn this paper, the traffic accident data was described, analyzed and extracted. The association rules was got successfully, which further prove that the extraction technology of association rules was valuable...