Robust Feature Selection Using Rough Set-Based Ant-Lion Optimizer for Data Classification

The selection of an algorithm to tackle a certain problem is a vital undertaking that necessitates both time and knowledge. Non-functional needs, such as the size, quality, and nature of the data, must frequently be taken into account. To develop a generalized machine learning model for any domain, the most relevant features must be chosen because noisy and irrelevant characteristics degrade data mining performance. However, the selection of the dominating features is still dependent on the search technique. When there are a high number of input features, stochastic optimization can be applied to the search space. In this research, we investigate the Ant Lion Optimization (ALO), a nature-inspired algorithm that mimics the hunting process of ant lions and is further stimulated to identify the smallest reducts. We also investigate Rough Set based ant lion optimizer for feature selection. The actual results reveal that the antlion-based rough set reduct selects a better feature subset and classifies them more accurately.