Safe Driving Based On Granular Computing Data Mining

With the approach of automotive society, the increasing number of motor vehicles, the number of motor vehicle drivers are the rapid growth of traffic safety has increasingly become an important factor of social development. Database technology in traffic management in general has received the application, after several decades of development, traffic safety, the database system has saved a lot of business data, how to dig a large database system that useful knowledge for the potential traffic management to provide guidance to become a pressing issue.Data mining is from a large, noisy, incomplete, and random, the practical application of fuzzy data, extraction of people do not know in advance, in which implied, but is potentially useful knowledge and information processes. Association rule mining is used to detect or interesting associations between large amounts of data or relationship. Data mining association rules found from the Traffic Safety database of potential and useful rule to guide the traffic management and reduce accidents. Classical algorithm of association rules APriori algorithm for pattern matching so many times as necessary to access the database frequently, thus greatly reducing its speed. The granular computing is a large complex problem is divided into several small problems easily, and its basic idea is on a different level of granularity problem solving, is to study the structure of multi-level thinking, problem solving methods, Information processing models and theories, techniques and tools of the subjects. In this paper, the particle thought to improve the calculation of association rules mining algorithm, and applied to traffic safety management, better knowledge of mining to improve efficiency and accuracy.On the basis of the above, careful analysis of road traffic management in the cars, people, environment, way of various factors on the formation of traffic safety and traffic safety data with the characteristics of road traffic safety attributes build the data model, traffic safety Data pre-processing. ORACLE database and then use the database layer, implemented a Visual Basic environment applications. Design the overall framework of data mining using granular computing and data mining correlation algorithm the steps to achieve traffic safety data mining.Proposed based on granular computing using multi-dimensional association rules algorithm to Luliang January 1,2004-2009 on December 31 for the study of traffic safety data objects, data processing and analysis of traffic accidents, and the association rule mining, sup and con have been associated with the rules of the system to achieve the results were analyzed by the rules can lead to good research the root causes of traffic accidents, traffic management and relevant departments for the proposed program of practical value, and gives for the road Traffic safety proposals for providing guidance to prevent traffic accidents.