A New Rough set-based Heuristic Algorithm for Attribute Reduct
Zhiqiang Geng, Qun-Xiong Zhu · 2006
Learning algorithms of data mining are known to degrade in performance when faced with many attributes that are not necessary for rule discovery. Rough set theory has been a topic of general interest in the field of knowledge discovery. A new rough set-based greedy heuristic algorithm is proposed for attributes reduct and emphasized the role of basic constructs of rough set approach. The approach can select an optimal subset of attributes quickly and effectively from a large database with a lot of attributes. So the sensitivity of rough set to noise can be depressed and the system's robustness is to be improved. The validity of the proposed algorithms is verified by comparing with genetic algorithms, johnson's algorithm and dynamic reducts in using practical machine learning databases.