An Oversampling Mechanism for Multimajority Datasets using SMOTE and Darwinian Particle Swarm Optimisation

Data skewness continues to be one of the leading factors which adversely impacts machine learning algorithms performance. An approach reduce this negative effect data variance is pre-process former dataset with level resampling strategies. Resampling strategies have been seen in two forms, oversampling and undersampling. strategy proposed article for tackling multiclass imbalanced datasets. This optimises state-of-the-art technique SMOTE Darwinian Particle Swarm Optimization technique. method DOSMOTE generates synthetic optimised samples balancing will more effective on multimajority experimental study performed peculiar datasets measure effectiveness approach. As a result, produces promising results when compared conventional