Association Rule Mining from Large and Heterogeneous Databases with Uncertain Data using Genetic Network Programming

Association Rule Mining is one of the most important tasks in data mining and it has been deeply studied during last years. Recently several rule mining algorithms have been developed due to many real-world applications. Most of these studies have generally considered only precise data, which means that items within each datum or transaction are definitely known and precise. However, there are also many real life situations where the data is uncertain, which means that items are expressed in terms of existential probabilities. In this paper, a method for association rule mining from large, heterogeneous and uncertain databases is proposed using an evolutionary method named Genetic Network Programming (GNP). Some other association rule mining methods can not handle uncertain data directly, they are inapplicable or computational inefficient under such a model. GNP uses direct graph structure and is able to extract rules without generating frequent itemsets to improve mining efficiency. The performance of the method is evaluated through extensive experiments using real scientific large-scale heterogeneous databases that show its effectiveness and efficiency. Keywords-Association rule mining; heterogeneous databases; uncertain data; evolutionary computation.