A Study On Association Rules Mining Algorithm And Its Application

Data Minging is a most active and new research in the field of Artificial Intelligence and field of database in the world.It's goal is to satisfy the users' need, process the great amount of original data and identify the important and interesting patterns in It.And at last expressing them In the form of knowledge which people can recognize at their first sight. Association rules are an important aspect of research of Data Minging. It has become a hotspot in data mining research field that mining association rules from large database, it has make delightful progress and explored many characteristic system info.This article introduced the definition and the main technologies of Data Mining at first, then described the association rules mining theory and the algorithm in detail, including classical Apriori algorithm and FP-Growth algorithm, and optimize the Apriori algorithm. In order to validate the validity of the optimizing algorithm in this article, the article used the supermarket transaction data to carry on in the efficiency analysis and the comparison, the experimental result has proven the validity of optimized algorithm. Finally, this article did some research in the application that using association rules mining in the fault analysis of HaFei Motor Co.Ltd, designed and realized a platform of association rules mining.The main work of this article represents in following several aspects:The first, this article had made the comprehensive analysis to the classics Apriori algorithm. Aiming at the deficiency of Apriori algorithm, this article has adopted one optimized method that using the method of reducing business data quantity in the discovery frequent item of collection database useless business records. Base on this optimization, the article adopted one kind method based on temporarily table, this algorithm holds the superiority in the data scale and the consumption of space, sped up the algorithm to efficiency. Finally, did the experiment using the data of supermarket, carried on the contrast to the operating efficiency of Apriori algorithm, temporarily table algorithm and Incremental Updating algorithm transferring temporarily table algorithm, which...