ABSTRACT
Machine learning is the process of teaching computers to learn from data and think like humans. Ensemble learning is a process by which multiple machine learning models can be combined strategically to solve a problem. In this subject, a new approach is proposed to find the optimal weights for a weighted ensemble classifier. The optimised weights can be found using an optimiser. Normally, grid search is used to find the optimal weights for the ensemble model. This subject proposes a new approach, Optimised weighted ensemble classifier (OWEC), to find the optimal weights for a weighted ensemble classifier using particle swarm optimisation as an optimiser. This approach is shown to be able to generalise to different datasets with different base learners, reduce the bias and variance of the overall model and is computationally more efficient to train compared to the base learners.
