Rough set theory demonstrated as a successful tool for data reduction in the last few years. Fuzzy rough set discarded the limitations of rough set theory to handle realvalued datasets. Intuitionistic fuzzy rough set was incorporated as an interesting tool to cope with the vagueness and uncertainty in a much better way when compared to fuzzy set based approaches specifically in the scenario where uncertainty was inserted not only by judgement but also by identification. In this paper, we establish a new intuitionistic fuzzy rough set model based on the concept of entropy to deal with imbalanced dataset