This project explores skin cancer detection using the SIIM-ISIC Melanoma Classification dataset, featuring 33,126 annotated skin lesion images. Employing a range of models including Decision Trees, Random Forests, Linear Regression, CNNs, and RNNs, we aim to optimize predictive performance while integrating demographic data, exploring image segmentation, and unsupervised learning. Our approach begins with thorough exploratory data analysis to inform model development. By rigorously evaluating models and documenting findings, we strive to advance computer-aided diagnosis capabilities, ultimately contributing to improved healthcare outcomes and awareness of machine learning's potential in healthcare.