A Time-Reduction Strategy to Feature Selection in Rough Set Theory

摘要

In rough set theory, the problem of feature selection aims to retain the discriminatory power of original features. Many feature selection algorithms have been proposed, however, quite often, these methods are computationally time-consuming. To overcome this shortcoming, we introduce a time-reduction strategy, which can be used to accelerate a heuristic process of feature selection. Based on the proposed strategy, a modified feature selection algorithm is designed. Experiments show that this modified algorithm outperforms its original counterpart. It is worth noting that the performance of the modified algorithm becomes more visible when dealing with larger data sets.