Evaluating and Enhancing the Performance of Skin Disease Classification Based on Ensemble Methods

Nowadays, people of all ages are suffering from various diseases. Skin cancer is one of the most common problem and it seriously impacts on human life and health. Fungal infection, bacteria, allergy, harmful UV rays of sunburn, etc. may be the reason for various types of skin disease. It can be recover if the disease can diagnose more quickly and correctly in the novel stage. At present, Machine Learning (ML) plays a significant role in the medical field. So, ML techniques helps to increase the accuracy in diagnosis of several types of skin disease. In our study, for evaluating the performance we have used five Machine Learning techniques like Naïve Bayes (NB), K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Random Forest (RF) and Multilayer Perception (MLP). We have also used three basic ensemble methods as Bagging, Boosting, Stacking to enhance the performance of the trained model and attaching all classifiers in the classification stage. Accuracy, Precision, Sensitivity, f1-score are considered as a measure of the usefulness of the proposed model. In this paper, our main motive is to progress the performance of the diagnosis of variant type of skin diseases and compare three ensemble techniques by combining all machine learning approaches.