Unsupervised feature selection using rough set and teaching learning-based optimisation

Feature selection is a valuable technique in data analysis for information preserving data reduction. This paper proposes to consider an information system without any decision attribute. The proposal is useful when we get unlabeled data, which contains only input information condition attributes but without decision class attribute. TLBO clustering algorithm is applied to cluster the given information. Decision table could be formulated using this clustered data as the decision variable. Then rough set and TLBO algorithms are applied for selecting features. The experiments are carried out on datasets of UCI machine repository and from the website http://www.ailab.si/orange/datasets.asp to analyse the performance study of our proposed approach with other approaches like genetic algorithm, particle swarm optimisation and differential evolution techniques. The results clearly reveal that our proposed approach outperforms other approaches investigated in this paper.