A New Approach to Mine Frequent Itemsets

Mining frequent patterns in transaction databases and many other kinds of databases has been studied popularly in data mining research. Methods for efficient mining of frequent itemsets have been studied extensively by many researchers. However, the previously proposed methods still encounter some performance bottlenecks when mining databases with different data characteristics. The time required for generating frequent itemsets plays an important role. And also the poor efficiency of counting candidate itemset’s support count. In this study, we propose a new frequent itemsets tree (FI-tree) structure, which is used for storing frequent itemsets and their Tid sets. A distinct feature of this method is that it has runs fast in different data characteristics. Our study shows that a new approach has high performance in various kinds of data, outperforms the previously developed algorithms in different settings, and is highly scalable in mining different databases.