10Building Rule Base for Decision Making – A Fuzzy-Rough Approach
Sabu M. K.1*, Neeraj Krishna M. S.2 and Reshmi R.3
1 Department of Computer Applications, Cochin University of Science and Technology, Cochin, Kerala, India
2 C3m India Private Limited, Ayyappankavu, Kochi, Kerala, India
3 Department of Computer Science and Engineering, Federal Institute of Science and Technology, Mookkannoor, Kerala, India
Abstract
Models based on Fuzzy sets and Rough Sets are efficient in handling vagueness present in information systems. In Fuzzy Set Theory, to handle vagueness, a fuzzy membership function is applied, whereas in Rough Set Theory the equivalence classes of an equivalence relation are considered. Usually, Rough Set methods were inefficient to process real-valued features effectively. To address this drawback, in the proposed system, a data discretization using Fuzzy Set Theory is performed to assign linguistic labels to the feature values. This is typically implemented with the help of a standard fuzzification technique provided by the Fuzzy Set Theory. This paper conducts a novel fuzzy rough approach for constructing rule bases for decision making. In this approach, after performing the fuzzy discretization, decision rules are generated with the help of a popular rough set-based rule mining algorithm. To explain the process of rule induction, an experiment is conducted to predict the yield of coconut using the data collected from farmers cultivating coconuts. The generated rules ...
Get Data Engineering and Data Science now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
