NCFP-tree: A Non-Recursive Approach to CFP-tree using Single Conditional Database
R. Prabamanieswari · International Journal for Research in Applied Science and Engineering Technology · 2017
The CFP-tree is one of the efficient FP-tree based mining algorithms to find frequent itemsets. It stores all frequent item sets in compact form. It is more disk-friendly than FP-tree. But, it creates multiple conditional databases during tree construction and it takes more I/O and CPU consumptions. It also has redundancy in creating the conditional databases. In order to avoid these problems, the non-recursive algorithm NCFPGEN for creating a NCFP-tree is proposed in this paper. The NCFP-tree is similar to CFP-tree but, it is created in non-recursive manner. The proposed algorithm creates an extended conditional database instead of creating multiple conditional databases in CFP-tree. The experimental results show that our method outperforms the existing method such as CFP-tree in both memory consumption and execution time aspects. This proposed NCFP-tree can be utilized in any frequent itemset mining based algorithms such as association rule mining, classification and representative patterns set generation. Keywords: FP-tree