Skin cancer is a common and deadly disease, and melanoma is the most lethal type. Early diagnosis is crucial in enhancing the treatment results. In this research, a cloud-based Convolutional Neural Network (CNN) system for skin cancer detection and classification was suggested. It uses the ISIC Skin Cancer Dataset, which consists of images with high resolution, and they are subjected to preprocessing techniques like image augmentation, normalization, and resizing. The CNN model is also optimized using the Adam optimizer to achieve efficient training and classification results. The effectiveness of the proposed model is validated through comparison with other methods, such as Hybrid GBDT+ALBERT+Firefly, CART+PLS-SEM, and PSPNET-HHT-Fuzzy Logic. It is revealed through results that the proposed CNN has better performance, with accuracy of 99%, precision of 96%, recall of 97%, and F1-score of 95.16%. The cloud infrastructure provides scalable storage and accessibility of large image datasets, promoting collaboration in health research. The method shows promising potential for clinical use in early skin cancer detection.