Exploratory data analysis was performed on a heart disease dataset, constructing various visualisations, to gain useful insights about the data. Various machine learning algorithms were then trained to classify whether someone has heart disease or not based on various features, scoring each algorithm based on their accuracy and deciding which algorithm was the best to use for further predictions. These algorithms included logisitic regression, K nearest neighbour, decision trees and random forest, of which t was found that the K nearest neighbour algorithm provided the most accurate predictions.