Artificial neural network may perform good to predict the survivability of cervical cancer

The project demonstrated to analyze the survivability of cervical cancer from the large data set of SEER (Surveillance Epidemiology and End Results). The data were re-sampled into 10 folds on 5 different size that were based on for three methods- artificial neural network, logistic regression and decision tree- to establish models for predicting the survivability of cervical cancer. In the meanwhile, 10-fold cross-validation was used to examine the respective models of three methods for performance comparison.