Research On Association Rule Mining Algorithm

Data Mining is one of the most active research fields, especially in the fields of artificial intelligence and Database. Data Mining is a kind of process that reveals potential useful knowledge from massive data. The association rule mining is a main research aspect of data mining. And the discovery of the frequent item sets is a key problem of the association rule mining algorithm.Firstly, the Data Mining is discussed generally in this paper,including it's concepts, patterns, applications and development trend. Secondly, Apriori algorithm is researched and analysed deeply, which is classic of the association rule mining algorithms. And then summarizes the problems existing in Apriori algorithm. For improving the existing poor efficiency of Apriori algorithm, this paper expounds some schemes in two aspects. One sheme is reducing the candidate item sets, and the other is decreasing the times spending in scanning Database. Then, a more efficient algorithm for association rule mining is presented. The main idea of the algoritnm is to mark all the transactions supporting each item in scanning Databese. The new algorithm adopts a unique way to generate the frequent item sets. It can mine all the frequent item sets by scanning the source Database only once and reducing the candidate item sets depending on the pruning. It not only decreases the bear of I/O, but also reduces the execution time, and then got high efficiency. Finally, the new method based on association rules is applied into the teaching evaluation, the mining result is analyzed, and instructive opinion is proposed.Although the certain results has been obtained in this paper, there are still many questions to be studied hardly, such as, designing the association rule mining application system, studying the interesting association rules and presenting the mining results more friendly to the user.