Cancer is a big issue in the whole world. It has many subtypes, which includes Blood cancer, Skin cancer, Lung cancer, Breast cancer, etc. Breast cancer is one of the most leading causes of death among women. The factors that cause this disease cannot be easily determined. The early detection of abnormalities in breast enables the doctor to treat the breast cancer easily. The diagnosis process which determines whether the cancer is benign or malignant also requires a great deal of effort from the doctors and physicians. A variety of Machine learning algorithms have now been applied to detect breast cancer, which includes Artificial Neural Networks (ANN), Bayesian Belief Networks (BBN), Support Vector Machines (SVM) and Decision Tree (DT) [1]. Many research papers about classification of breast cancer have only considered two classifiers such as a high and low-risk group. But, the binary classification detects cancer at the later stages, which is difficult to cure and the other drawback is it is error-prone i.e., the results of binary classification are not accurate. The error rate can be still decreased by multi-classifying the cancer data. The various Multi-class classification algorithms are Neural Networks, K-Nearest Neighbors, Boosting, Decision Trees etc. In this work, the three algorithms SVM, KNN, Gaussian Naïve Bayes algorithms are used for classification and K-means algorithm is used for clustering. The performance of these algorithms is analyzed.