Prediction of the success of health therapy using machine learning

Thesis Type Bachelor
Thesis Status
Finished
Student Samuel Plangger
Init
Start
Thesis Supervisor
Contact

In Austria electronic health records have been stored since 2013. This data has proven to be a very useful tool for future patient care.

The goal of this thesis is to predict the therapy success of patients. To achieve this a dataset consisting of about 2500 patients was used. The predictions based on the dataset are made with three machine learning algorithms. In the preparation process the dataset was prepared for use with machine learning algorithms and the machine learning algorithms have been optimised for the dataset. The result of this thesis is that all used algorithms exhibit a prediction error of less than 10%.