Home // DATA ANALYTICS 2012, The First International Conference on Data Analytics // View article

A New Measure of Rule Importance Using Hellinger Divergence

Authors:
Chang-Hwan Lee

Keywords: Rule Importance; Association; Information Theory;

Abstract:
Many rule induction algorithms generate a large number of rules in data mining problem, which makes it difficult for the user to analyze them. Thus, it is important to establish some numerical importance measure for rules, which can help users to sort the discovered rules. In this paper, we propose a new rule importance measure, called HD measure, using information theory. A number of properties of the new measure are analyzed.

Pages: 103 to 106

Copyright: Copyright (c) IARIA, 2012

Publication date: September 23, 2012

Published in: conference

ISSN: 2308-4464

ISBN: 978-1-61208-242-4

Location: Barcelona, Spain

Dates: from September 23, 2012 to September 28, 2012