Re-heat simulated annealing algorithm for rough set attribute reduction
Salwani Abdullah, Laleh Golafshan, Mohd Zakree Ahmad Nazri · 2011
Attribute reduction deals in finding all possible reducts by egesting redundant attributes while maintaining the information of the problem in hand. This paper aims to investigate an alternative approach to find the minimal attribute from a large set of attributes. Towards this goal, a reheat simulated annealing (Reheat-SA) is proposed to solve an attribute reduction problem in rough set theory. It is a meta-heuristic approach that has a mechanism to escape from local optima. The concept of re-heat is introduced to help the algorithm to better explore the search space to finding a better solution. The proposed approach is tested on a well known UCI datasets. Results show that our approach is able to find competitive results when compared to state-of-the-art approaches. Key words: Re-heat simulated annealing, rough set theory, attribute reduction.