On the Efficiency of PCA-SVM Based Binary Skin Cancer Classification Technique

Melanoma, a deadly form of skin cancer, demands early detection for appropriate treatment. This research explores the potential of artificial intelligence (AI) and machine learning (ML) techniques in aiding early diagnosis. Specifically, the application of Support Vector Machines (SVM) for the classification of malignant and benign skin lesions has been performed in this paper. Principal component analysis (PCA) reduces the dimensionality of the feature space to enhance model performance and computational efficiency. A comprehensive dataset of dermoscopic images is utilized to train and evaluate the proposed SVM model. The model's ability to discriminate between malignant and benign lesions is assessed through metrics like confusion matrix and accuracy. The impact of feature reduction using PCA on classification accuracy is analyzed. This research contributes to developing innovative solutions for early skin cancer detection, potentially saving lives.