ABSTRACT

Treating missing value is very big task in the data preprocessing methods. Missing data are a potential source of bias when analyzing clinical trials. In this paper we analyze the performance of different data imputation methods in a task where the aim is to predict the probability of survival of cardiac patient. In this paper, comparison of handling missing data in cardiac dataset. Mean Imputation, KNN imputation method, two correlation based methods known as EMImputed _ columns, LSImputed _ Rows and multiple imputation method referred as NORM (which is based on Expectation Maximization algorithm) method were used to replace missing values found in a dataset containing 3500 records of patients. The results were analyzed in terms of the calibration of the results. Nevertheless, k-NN methods may be useful to provide relatively accurate estimations with lower error variability