A New Selection Measure for Classification Using Decision Trees

B. Chandra () and Gaurav Saxena
Additional contact information
B. Chandra: Indian Institute of Technology, New Delhi, India
Gaurav Saxena: Indian Institute of Technology, New Delhi, India

Journal of Information & Knowledge Management (JIKM), 2004, vol. 03, issue 01, 1-7

Abstract: The paper proposes a new selection measure for classification using decision trees for Data mining. Various algorithms have been proposed in the past for classification using decision trees viz. ID3, CART, SLIQ, etc. Selection measures like the Gain, Gain ratio, and Gini index have been proposed in these algorithms. However, none of the selection measures developed so far take into account the balancing of trees. This paper proposes a new selection measure which also takes into account the balancing of trees that will facilitate in improving the classification accuracy. The performance of the original SLIQ algorithm, C5 and the algorithm using the new selection measure (which takes into account the accuracy as well as the balance factor) was measured on the basis of the classification accuracy. Three real life data sets were chosen for this purpose.

Keywords: ID3; CART; SLIQ; Gini index; Gain ratio (search for similar items in EconPapers)
Date: 2004
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.worldscientific.com/doi/abs/10.1142/S0219649204000626
Access to full text is restricted to subscribers

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:wsi:jikmxx:v:03:y:2004:i:01:n:s0219649204000626

Ordering information: This journal article can be ordered from

DOI: 10.1142/S0219649204000626

Access Statistics for this article

Journal of Information & Knowledge Management (JIKM) is currently edited by Professor Suliman Hawamdeh

More articles in Journal of Information & Knowledge Management (JIKM) from World Scientific Publishing Co. Pte. Ltd.
Bibliographic data for series maintained by Tai Tone Lim ().