A primary study for cancer prognosis based on classification and regression using support vector machine

Abstract

In medical domain, prognosis prediction treated as a regression problem is generally applied to predict the event duration time, such as the duration time of the recurrence of a certain disease. Recently, machine learning techniques are gaining popularity in this field because of its effectiveness and reliability. In this paper, a method based on support vector machine (SVM) to predict the exact recurrence time has been proposed. The method is compared with other four prognostic methods using Wisconsin Breast Cancer Dataset. Experimental results demonstrate that the method is more simplified to be implemented than the other four prognostic methods, and it performs much better than the medium level. © Springer Science+Business Media Dordrecht 2014.