Computational intelligence and hybrid models for feature selection and classification of bioinformatics datasets

摘要

The science of bioinformatics is increasingly being used to improve the quality of life as we know it. This opens the door to both machine learning and computational intelligence techniques to play a major role in bioinformatics. Recently, Type-1 and Ttype-2 fuzzy logic systems have been introduced as novel computational intelligence approaches for both prediction and classification. They were successfully used in several areas of science and engineering, however, they have not been fully utilized in the bioinformatics, particularly the Type-2 fuzzy logic system. This paper presents various computational intelligence and hybrid models for feature selection and classification of real bioinformatics datasets. The performance and classification accuracy of the presented models are measured using well known bioinformatics datasetsfrom the machine learning repository at University of California Irvine. Empirical results have shown that the proposed hybrid models outperform earlier models with better classification accuracy.