Abstract—Two classical measures, approximation accuracy and consistency degree can be employed to evaluate the decision performance of a decision table. However, these two measures cannot give elaborate depictions of certainty and consistency of a decision table when their values equal to zero. Yuhua Qian et al have introduced three new measures for evaluating the decision performance of a decision–rule set extracted from a decision table, but these measures still have some limitations. In this paper, we propose a new measure and theoretically study to evaluate the consistency of a set of decision rules extracted from a decision table. Experimental analyses on two practical data sets also show our measure will be suited for evaluating the decision performance of a decision-rule than measures of Yuhua Qian et al.

Index Terms—accuracy degree, consistency degree, decision table, decision evaluation, rough set theory

Nguyen Duc Thuan is with the Department of Information Systems, Nha Trang University, Vietnam (phone: +84 58 905444448; fax: +84 58 3831147).
Nguyen Xuan Huy is with the Institute of Information Technology Vietnamese Academy of Science and Technology, Hanoi, Vietnam

Cite: Nguyen Duc Thuan and Nguyen Xuan Huy, "A New Measure to Evaluate the Consistency of a Set of Decision Rules Extracted From a Decision Table," International Journal of Computer and Electrical Engineering vol. 1, no. 4, pp. 431-435, 2009.