Performance Study of Graph Convolutional Networks for Medical Prediction-Based Networks

Predicting the effects of Polypharmacy is a difficult task, and a great amount of money is spent annually remedying the effects of negative drug interactions arising from Polypharmacy. However, Machine Learning can be used to give more accurate predictions than traditional means. In this thesis, we survey current methods of applying Machine Learning to Polypharmacy. We rigorously define a theoretical Polypharmacy problem and design a Graph Convolutional Network that can learn to strongly model our problem. We discuss its performance and offer future steps for generalizing the model to gain a better understanding of the field of Polypharmacy and the potential of Machine Learning to improve it