Hotspots are residues which make dominant contributions to the free energy of binding at protein interfaces. Experimentally, a hotspot can be identified by mutating it to alanine and measuring the changes in free energy of binding (ΔΔG). Experimental information is available only for a limited number of complexes. Hence, a need for computational methods arises. Several methods based on machine learning algorithms are implemented to predict hot spots. Furthermore, sequence and/or structure based features are used for determining whether a residue at protein interface is a hotspot or not. Additionally, a lot of data sets are used but some of them are redundant or incommensurate in the context of hotspots. In this study, we offer a critical assessment of the methods and features used for hot spot residue prediction at the protein-protein interface in recent years and also propose a newly generated non-redundant protein data set.