This article provides an outline on a recent application of soft computing for the mining of microarray gene expres-sions. We describe investigations with an evolutionary-rough feature selection algorithm for feature selection and classifica-tion on cancer data. Rough set theory is employed to generate reducts, which represent the minimal sets of non-redundant eatures capable of discerning between all objects, in a multi-objective framework. The experimental results demonstrate the ffectiveness of the methodology on three cancer datasets.
展开▼