Improvement in Performance of Classificstion Using Lift Method

Author(s): Neha P. sonar, Ankit K. Dixit, Kajal E. Shelke, Viren A. Patil, Rahul N. Patil

Publication #: IJIRCT1601038

Date of Publication: 15.04.2017

Country: India

Pages: 205-207

Published In: Volume 2 Issue 6 April-2017

Abstract

Data mining is the process of analyzing data from different perspective and summarizing it into useful information. Classification is one of the important task in data mining. Interestingness of pattern in a given class is measured by the two terms- support and cohesion. The set of classification rules are generated from these interesting pattern. In existing system these set of classification rules are used to classify the dataset. But the set of rules may contain independent rules or rules which are not correlate to each other. Such independent rules increases classification time. So, there is need to prune such independent rules, for which lift technique is used. The lift technique will reduce the amount of time require for classification of data.

Keywords: Interesting pattern, Classification rules, lift Method, Classification

Download/View Paper's PDF

Download/View Count: 383

Share this Article