Application of Ensemble Machine Learning for Classification Problems on Very Small Datasets

Machine learning is one of the most widely used branches of artificial intelligence in recent years. It is most commonly used for solving classification or regression problems through the utilization of supervised learning approaches. Machine learning models require high quality and a sufficient quantity of data to produce good results. This paper investigates an approach which incorporates ensemble learning through the aggregation of multiple machine learning models for the purposes of increasing prediction capabilities in cases in which a very limited amount of data is available for training. The ensemble model was trained on a patient fractional flow reserve biomarker dataset and with the goal of classifying patients into risk classes based on their risk of suffering an acute myocardial infarction. The ensemble model was comprised of multiple random forest classification models which were trained with different combinations of training and test data to improve the prediction accuracy over the use of a single random forest model. Final ensemble achieved a prediction accuracy of 71.3% which was an immense improvement over the 36% prediction accuracy of a single random forest classification model.