Synergistic Combination of Machine Learning and Evolutionary and Heuristic Algorithms for Handling Imbalance in Biological and Biomedical Datasets

Due to the advent of Next Generation Sequencing and multiple innovative experimental techniques there is an exponential increase in biological data. It is necessary to capture meaningful information and valuable knowledge from this data which can immensely benefit all living things on the planet. Recent advances in machine learning and deep learning have indeed been able to handle this with the deployment of novel algorithms. Data imbalance frequently occurs in biological data mining tasks. This is because in several omics related problems data belonging to the positive class is much less than the data belonging to the negative class. Such imbalance can affect performance and produce faulty results and conclusions. Nature inspired Evolutionary and Metaheuristic algorithms are robust and can handle data imbalance quite efficiently. In this work we have described the use of these algorithms for tackling data imbalance in biological data. We have provided lucid explanations of these algorithms along with potentially important case studies.