MCFSP: Mining Closed frequent Similar Patterns

Computer Science

The enormous amount of useful information in the form of frequent pattern poses the key challenge of how to reduce the number of frequent patterns without information loss. Despite the quality of solution given by frequent pattern mining algorithm, scalability and efficiency issues persist. Current closed frequent item set mining algorithms gives solutions only to frequent itemset by discovering reduced sets of frequent itemsets from which entire frequent itemsets can be recovered . But the solution in the case of frequent similar pattern mining wherein the number of pattern is even more than for frequent itemset mining is not present. In this paper as a solution we are extending closed frequent pattern mining technique to closed frequent similar pattern mining for discovering reduced set of frequent similar pattern without any information loss. We are proposing a novel closed frequent similar pattern mining algorithm, named MCFSP Mining closed frequent similar pattern algorithm. By traversing a tree which contains all closed frequent similar pattern the algorithm discovers frequent patterns. The MCFSP resolves scalability and efficiency issues by finding closed similar patterns, yielding a reduced size of the discovered frequent similar pattern set without information loss.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.