posted on 2017-12-06, 00:00authored byMohammed Mazid, A B M Shawkat Ali, Kevin Tickle
C4.5 is one of the most popular algorithms for rule base classification. There are many empirical features in this algorithm such as continuous number categorization, missing value handling, etc. However in many cases it takes more processing time and provides less accuracy rate for correctly classified instances. On the other hand, a large dataset might contain hundreds of attributes. We need to choose most related attributes among them to perform higher accuracy using C4.5. It is also a difficult task to choose a proper algorithm to perform efficient and perfect classification. With our proposed method, we select the most relevant attributes from a dataset by reducing input space and simultaneously improve the performance of this algorithm. The improved performance is measured based on better accuracy and less computational complexity. We measure Entropy of Information Theory to identify the central attribute for a dataset. Then apply correlation coefficient measure namely, Pearson’s, Spearman, Kendall correlation utilizing the central attribute of the same dataset. We conduct a comparative study using these three most popular correlation coefficient measures to choose the best method on eight well known data mining problem from UCI (University of California Irvine) data repository. We use box plot to compare experimental results. Our proposed method shows better performance in most of the individual experiment.
Funding
Category 4 - CRC Research Income
History
Start Page
296
End Page
301
Number of Pages
6
Start Date
2010-01-01
ISSN
1790-5109
ISBN-13
9789604741540
Location
University of Cambridge, Cambridge, UK
Publisher
WSEAS Press
Place of Publication
London
Peer Reviewed
Yes
Open Access
No
External Author Affiliations
Faculty of Arts, Business, Informatics and Education; Institute for Resource Industries and Sustainability (IRIS);
Era Eligible
Yes
Name of Conference
International Conference on Artificial Intelligence, Knowledge Engineering and Data Bases