A.Rayudu, B.Ashwini, B.Asritha, A.Umashankar, Mrs.Kalpana,

Abstract

Because few people know the warning symptoms of skin cancer and how to avoid getting it, it ranks high among the most hazardous illnesses. The shockingly high mortality toll from skin cancer has led some to term it a "fourth burden disease" on a global scale. As a result, cancer cannot progress beyond its first stages unless it is detected early. Using ML and IR methods, we identify and categorize multi-label skin cancer and use the best practices in this research. Nevertheless, preprocessing techniques help eliminate superfluous and irrelevant features from the label encoder, and standard features are used to scale the input variance unit and standardize the range of functionality. In addition, the HAM10000 metadata dataset was used to evaluate each classifier's performance using a variety of machine learning approaches. The seven distinct skin cancer types included in the HAM10000 metadata dataset were the subjects of the experimental study. According to the findings, SVM, DT