Skin Disease Classification using Machine Learning based Proposed Ensemble Model

Abstract

Skin disease is a major issue of global health problem affecting a large amount of persons. The advancement of dermatological diseases categorization has grown more accurate in recent years due to the rapid growth of technology and the use of various machine learning techniques. Therefore the creation of machine learning methods that can accurately differentiate between the classifications of skin diseases is one of the great importance. This research work focuses on the classification of different kinds of skin diseases using machine learning techniques. In this research, we introduce a novel approach that makes use of four distinct data mining techniques like support vector machine (SVM), k-nearest neighbor (KNN), random forest (RF) and, naive bayes (NB) algorithm. This research work proposed an ensemble model that is combination of SVM, KNN, RF and NB using voting scheme. The proposed model classified the skin disease into five different classes that are Acne, Skin allergy, Nail fungus, Hair loss, and Normal skin. The proposed ensemble model used on skin disease classification that gives better performance over other classifier algorithms. The proposed ensemble model achieved highest 97.33% of accuracy as compared to others.