The model is optimized using Bayesian hyperparameter optimization. Particle Swarm Optimization (PSO) technique may also be used for parameter optimization. PSO considers the past evaluation results to choose the next values to evaluate and it may be faster than the Bayesian technique. The proposed deep learning model gives a good result so far. Research shows that the hybrid ensemble techniques also effective for image analysis. The evaluation metrics accuracy, precession, and sensitivity give a good indication that the model in this research is promising. Since the original data is imbalanced explicitly stating the F1 score would be useful.
Selvarajah Mohanarajah PhD
Chair & Associate Professor of Computer Science Department of Mathematics and Computer Science University of North Carolina at Pembroke firstname.lastname@example.org
Safaa Alwajidi PhD
ML Data Scientist & Associate Professor of Computer Science Department of Mathematics and Computer Science University of North Carolina at Pembroke email@example.com