New adaptations of classic algorithm for mining frequent itemsets from uncertain data

Mining frequent itemsets from traditional database is an important research topic in data mining and researchers achieved tremendous progress in this field. However, with the emergence of new applications, the traditional way of mining frequent itemsets is not available in uncertain environment. In the past ten years, researchers proposed different solutions in extending the conventional techniques into uncertainty environment. In this paper, we review the previous algorithms based on the two definitions of frequent itemsets, then we improve the traditional classic algorithm for mining frequent itemsets in uncertain databases under the definition of frequent probability. Finally, we tested our algorithm on a number of uncertain data sets. The experiments on both sythetic and real data have shown that the new adaptation of classic algorithm is efficient and gain better results on accuracy.