Research Article 
Measuring the Interestingness of Classification Rules
Sanjeev Sharma, Swati Khare and Sudhir Sharma
Asian Journal of Information Management, 2007, 1(2), 43-49.
Abstract
Data mining tools and techniques provide various applications with novel and significant knowledge. This knowledge can be leveraged to gain competitive advantage. However, the automated nature of data mining algorithms may result in a glut of patterns-the sheer numbers of which contribute to incomprehensibility. Importance of automated methods that address this immensity problem, particularly with respect to practical application of data mining results, cannot be overstated. We provide a survey of one important approach, namely interestingness measure and discuss its application to extract interesting results out of large number of rules generated by the classification rule generator program. We have used the US Census database of UCI repository as our experimental domain. Rules are generated by the Christian Borgelts classification rule discovery program. A new rule selection mechanism is introduced and experimental results show that our method is effective in finding interesting rules.
ASCI-ID: 24-7
Similar Articles
A Comprehensive Comparative Study Using Vector Space Model with K-Nearest Neighbor on Text Categorization Data
Asian Journal of Information Management, 2008, 2(1), 14-22.
Design of Algorithm for Frequent Pattern Discovery Using Lattice ApproachAsian Journal of Information Management, 2007, 1(1), 11-18.
Fast Algorithm for Mining Multi-Level Association Rules in Large DatabasesAsian Journal of Information Management, 2007, 1(1), 19-26.
Cited By
Extracting Association Rules from Hiv Infected Patients’ Treatment Dataset
Trends in Bioinformatics, 2011, 4(1), 35. DOI: 10.3923/tb.2011.35.46
Connection Subgraphs: A SurveyJournal of Applied Sciences, 2011, 11(17), 3221. DOI: 10.3923/jas.2011.3221.3232