Astrophysics inspired multi-objective approach for automatic clustering and feature selection in real-life environment

In this paper, a novel astrophysics-based approach is proposed for automatically finding the clusters and features simultaneously. A novel agent encoding scheme is used to encode both the number of clusters and features. A novel dynamic threshold technique is proposed for an efficient searching. The validation of proposed technique is tested on eight real-life data sets. The statistical significance of proposed technique is attributed by statistical tests. It is also applied on solving image segmentation and microarray data analysis problems. Experimental results reveal that the proposed technique outperforms the other competitive approaches.