Machine Learning Models in Predicting Hepatitis Survival Using Clinical Data (Extended Abstract)

In this study on the prediction of survival in hepatitis patients, the Decision Tree proved to be the most efficient classification model. Therefore, this model can be recommended as a useful tool for the prediction of survival of hepatitis patients as well as for medical decision making. This method is quick to apply because it is resistant to training and the ability to manage data with and without pre-processing, for example, the data does not need to be resized, transformed or modified. It can also be used for feature selection (searching for effective risk factors) only. The analysis of this study showed that the absence of variables such as steroid, antiviral, liver-big and liver-firm do not significantly affect the prediction result. However, the increase in our dataset data could allow our model to have better performance than those presented above.