Skin cancer, one of the most prevalent types of cancer in the world, has gradually on the rise for the past several decades. The VGG-19 deep convolutional neural network is used in this study to introduce a reliable method for the diagnosis and classification of skin cancer that is enhanced by transfer learning and data augmentation approaches. Further, we fine-tune the network to improve its diagnostic abilities using a pre-trained VGG-19 model on a substantial skin cancer image dataset. In addition, data augmentation technique is used to improve model generalisation and address the overfitting issue at the same time. Our approach exhibits astounding precision in the identification and classification of different forms of skin cancer, offering a viable tool for early diagnosis and accurate classification. This study makes a substantial contribution to the discipline of dermatology and offers an important resource for improving skin cancer diagnosis and patient care.