Evaluation of Dipolar Neural Networks in Survival Time Prediction

In this paper a dipolar neural network designed for prediction of survival time is presented. The proposed method is based on minimization of piece-wise linear criterion functions. Basis exchange algorithms are exploited as a optimization technique. The method allows to estimate of discrete hazards as conditional probabilities of failure occurrence. The prediction capacity of neural network models is evaluated by accuracy, sensitivity, and specificity measures, which are compared to results of logistic regression.