The unbalanced and inadequate distribution of medical services, for example, has grown more and more obvious in today’s climate, where medical care challenges have intensified. In this instance, the use of Machine Learning has been a natural progression in the development of modern medical care. As a result of the development of cosmetics use, numerous malignancies and tumors have appeared in the world in recent years. One such dangerous cancer or tumor is melanoma. The malignant illness known as melanoma occurs when the pigment-producing cells that give the skin its color turn cancerous. Signs could include a sudden, erratic growth or a change in an existing mole. Any area of the body might acquire melanoma. Hence Using methodologies involving lesion detection for accuracy, efficiency, and performance criteria, skin lesions are automatically detected. For the purpose of early skin lesion identification, the suggested approach employs feature extraction utilizing the ABCD (Asymmetry, Border, Color, Diameter) rule, GLCM (Gray Level Co-Occurrence Matrix, and HOG (Histogram Of Oriented Gradients) feature extraction. Pre-processing is used in the proposed work by researchers to enhance the skin lesion’s clarity and quality while reducing artefacts such skin color and hair. The lesion part was segmented individually using Geodesic Active Contour (GAC), which was also effective for feature extraction. To extract the properties of symmetry, border, color, and diameter, the ABCD scoring method was utilized. Textural characteristics were extracted using HOG and GLCM. Different machine learning approaches, such as SVM (Support Vector Machine), KNN (K-Nearest Neighbor), and Naive Bayes classifier, are used to categories skin lesions between benign and melanoma utilizing the retrieved features.