Improved Algorithm for Mining of High Utility patterns in one phase Based on Map Reduce Framework on Hadoop

Abstract

Mining high utility itemsets from a value-based
database alludes to the disclosure of itemsets with high utility like
benefits. In spite of the fact that various significant calculations
have been proposed lately, they bring about the problem of
causing a sizably voluminous number of applicant itemsets for
high utility itemsets. Such a large number of candidate itemsets
degrades the mining performance in terms of execution time
and space requirement. Earlier work shows this on two phase
candidate generation. This approach suffers from scalability issue
due to the huge number of candidates. Our paper presents the
efficient approach where we can generate high utility patterns
in one phase without generating candidates. Here we have
taken experiments on linear data structure, our pattern growth
approach is to search a reverse set enumeration tree and to prune
search space by utility upper bounding. Also high utility patterns
are identified by a closure property and singleton property. Iin
this venture we are displaying new approach which is extending
these calculations to conquer the restrictions utilizing the Map
Reduce structure on Hadoop. Experimental results show that the
proposed algorithms, not only reduce the number of candidates
effectively but also outperform other algorithms substantially in
terms of runtime, especially when databases contain lots of long
transactions.