Classification of Skin Cancer Disease Using Data Mining Techniques

Abstract

In the previous decade fast growing of digital facts and worldwide availability of it over current internet has perceived a huge increase in machine learning exploration. In ratio to that, the health data has similarly seen a huge range of development. By the obtainability of organized clinical data, it has involved researchers to learn on the computerization of medical disease discovery through machine learning and data mining. Melanoma is a fatal skin malignance that breakdowns in the skin’s tincture cells on the membrane shallow. Melanoma origins 75% of the skin cancer-associated deaths. This disease be able to identify by a dermatology expert over the clarification of the dermoscopy imageries in keeping with ABCD law. So, our investigation goals to study the automated discovery of skin cancer ailment through medical data by numerous machine learning classifier. This exploration mainly emphases on Neural Nets, Deep learning, Naïve Bayes, Random Forest classifier and decision tree in the determination of categorizing the intended dataset in three groups as normal, abnormal and melanoma to develop a decision support system that would create the assessment easier for a doctor. Generally our attempt has been to attain a supportable and realistic model to distinguish the skin cancer disease through comprehensive scientific accuracy.