Using support vector regression in gene selection and fuzzy rule generation for relapse time prediction of breast cancer

Gene expression profiles have been recently used in survival analysis, tumor classification and ER status identification. The prediction of breast cancer recurrence based on gene expression profile has been regarded in some previous studies in which the procedures were based on the concept of regression functions and fuzzy systems. In this study, a method based on the combination of these two concepts is presented; not only a method for gene selection, but also a systematic way to create fuzzy rules are going to be offered. Due to the ability of type-2 fuzzy systems in handling of uncertain systems, the proposed model is developed to type-2. The results show that this model has been improved in comparison to previous ones.