Mining Maximal Frequent Subtrees based on Fusion Compression and FP-tree

It is commonly accepted that mining frequent subtrees play pivotal roles in areas like Web log analysis, XML document analysis, semi-structured data analysis, as well as biometric information analysis, chemical compound structure analysis, etc. An improved algorithm, i.e. MFPTM algorithm, which based on fusion compression and FP-tree principle, was proposed in this paper to determine a better way to mine maximal frequent subtrees. The algorithm firstly retains subtrees which only contain frequent nodes by fusion compression, then according to FP-tree principle mines frequent subtrees. In the process of mining frequent subtrees, MFPTM algorithm is the means by which we attempt to satisfy our appetite for saving searching space of mining candidate patterns, and our craving to solve problems of frequent pattern mining based on Apriori algorithm which is generating a large quantity of candidate patterns. MFPTM algorithm, which actively represents as many viewpoints as is both possible and feasible as an advanced algorithm, improves the efficiency of mining frequent subtrees.