A critical challenge of the post-genomic era is to find out the cancer causing genes that induce changes in gene expression profiles of the microarray dataset. Suitable analysis of microarray dataset can unlock the mystery of the origin of many dreaded disease like cancer which can subsequently be investigated for its rectification, resulting into search for drug design. In this work, Rough Set Theory (RST)-based techniques have been applied for the same purpose. Rough Set has been used to classify the microarray data and isolate responsible genes from classification-based microarray dataset. This extracted information are essential for their ability to describe complex stochastic processes like gene transcription, classification of biological sequencing, and intuitive model for finding causal influence. The experiment is carried out on the microarray dataset of breast cancer (GSE-38867) which has four stages of cancer, i.e., normal, ductal, invasive, and metastatic. RST shows very promising results in terms of predicting different stages of cancer.