Improving Efficiency of MIKE Algorithm by Reducing Set Size

Abstract: Erasable item set mining was introduced to approach data mining in production planning. The erasable item set mining isthe process of finding erasable item sets that satisfy the constraint i.e. user defined value. This paper proposes an efficient algorithm for finding Top-Rank-K erasable item sets. Since the MIKE Algorithm was proposed to generate the top-rank-k erasable item sets. In last few years there have been several methods to improve its performance. But they do not consider the time and space constraint. If rank is high value then MIKE takes a lot of time and space to generate candidate set. In this paper, we proposed an Improved MIKE (I-MIKE) which reduces time and space by using efficient approach to generate candidate set.

Keywords: MIKE algorithm, data mining, Erasable item set, Apriori algorithm.