The Research And Implementation Of Association Rule Data Mining Algorithm

Data mining technique is an arisen research topic in database, artificial intelligence fields and so on, and it has attracted a great deal of attention in the information industry in recent years. Mining association rules from transaction database is most commonly seen in data mining. It discovers relations among items such that the presence of certain items in a transaction tends to imply the presence of certain other items. Association rules mining has become a top topics for its concise style and easy understanding. Now relational database is the most common means to store vast information of production, management and scientific research. Studying the efficient technology of association rules mining in relational database has a wide development in future.This paper analyzes Apriori algorithm, which is a classical algorithm about mining association rules in transaction database, and others. Then put forward an improved algorithm, CBApriori algorithm, according to some deficiencies of Apriori algorithm. It is core of the algorithm that original database is replaced by some cluster-tables for decreasing the times of scanning original huge database. At the same time, set multi-minsupport to deduct unnecessary items quickly. Thus the algorithm is improved efficiently.Based on the above theory research, we design and realize a primitive association rules mining tools model. And realize the visualization of mining results for users to understand it intuitively and quickly. With the development of data mining techniques, privacy protection and information security problem has emerged globally. We discuss methods of data mining based on privacy protection, and introduce a method of disguising data based on randomized response technique. And combine it with associaton mining algorithm.