Rough Set Based Ant-Lion Optimizer for Feature Selection
Ahmad Taher Azar, Nizar Banu, Anis Koubâa · 2020
As the area of computational intelligence evolves, the dimensionality of any sort of data gets expanded. To solve this issue, Rough Set Theory (RST) has been successfully used for finding reducts as it requires only supplied data and no additional information. This paper investigates a novel search strategy for minimal attribute reduction based on rough sets and Ant Lion Optimization (ALO). ALO is a nature-inspired algorithm that mimics the hunting mechanism of ant lions, and this is inspired to find the minimum reducts. Datasets from the UCI repository are used in this paper. The experimental results show that the features selected by the proposed method are well classified with reasonable accuracy.