SH-Struct: An Affirmative Advanced Method for Mining Frequent Patterns

Data Mining requires versatile computational techniques for analyzing patterns among large and diversified databases. One of the most influential and typically emerging research area is to develop impinging structures for valid frequent patterns. In this paper, we have formulated a novel data structure known as SH-Struct (Soft-Hyperlinked Structure) which mines the complete frequent itemset using SH-Mine algorithm. This algorithm enables frequent pattern mining with different supports. Fundamentally, SH-Struct is a tree structure which maintains H-Struct (Hyperlinked Structure) at each level of the tree called SH-Tree to improvise storage compression and reserves frequent patterns very fast using SH-Mine algorithm. To validate the effectiveness of our structure here we present the performance study of SH-mine and FP (Frequent Pattern)- growth algorithm highlighting space and time payoffs for two categories of databases: sparse and dense. The experimental results show the prominent behavior of proposed method and incite us to further deploy it in more dense and dynamic databases such as temporal databases for generating more prognostic outcomes.