Melanoma Prediction from Skin Lesions Using Convolutional Neural Network against the Support Vector Machine Learning Algorithm to Achieve Maximum Accuracy and Minimum Loss

Abstract

This chapter has the research study to predict the classification of melanoma from skin lesions using an innovative dermatoscopic image capture dataset using convolutional neural network (CNN) deep learning algorithms. This research aims to predict higher accuracy and lower loss of the classification of melanoma from skin lesion detection. This chapter used ten samples with two groups of algorithms with the g-power value of 80%, and the dermatoscopic images are collected from various web sources with recent study findings. To predict the classification of melanoma for humans, the support vector machine (SVM) machine learning algorithm has already found 81% of accuracy; therefore, this study needs to find better accuracy for skin lesion detection with the CNN deep learning algorithm. This chapter found 92% of accuracy for the classification of melanoma from skin lesion detection using the CNN algorithm. This study concludes that the CNN algorithm on the dermatoscopic innovative image capture dataset is significantly better than the SVM algorithm.