Memory boosting for pattern growth approach

The pattern growth approach of association rule mining is very efficient as avoiding the candidate generation step which is utilized in Apriori algorithm. This research is about the revisiting the pattern growth approaches to discover the different research works carried out to improve the performance using different criteria like header table dealing, item search order, conditional database representation, conditional database construction strategy and tree traversal strategy. And concluded that the reduction in overall time of pattern growth approach can be achieved by reducing the search space and processor operations time at the header table generation. It is proposed to achieve by only considering the items that are going to be frequent and ignoring the infrequent items at early stage of database scan, by considering the boundary. Compare to FP-Growth and CFP-Growth the proposed approach Modified FP-Growth (MFP-Growth) is outperforming in experimental analysis to achieve the cutback in execution time by utilizing the restriction on the memory allocation of infrequent items.