Unfortunately, around the world, breast cancer remains a major health risk for women. Early & accurate detection is a must. Our study has conducted evaluation of 6 different algorithms of machine learning, which are Random Forest, Naïve bayes, Logistic Regression, K-Nearest Neighbors, Decision Trees & Support Vector Machines. They are used for breast cancer classification, using data collected from breast cancer cases, that is well-known, which is Wisconsin Breast Cancer Diagnostic (WBCD) dataset. The models were assessed using several important performance metrics. The key performance indicators (KPIs) used in our study includes Recall, Accuracy, F1 Score & Precision. Among the six models, SVM achieved the highest accuracy of 95.62%, followed by Logistic Regression with 94.75%, Naive Bayes with 93.87%, KNN with 92.99%, Random Forest with 91.24% & Decision Tree with 87.73%. In conclusion, SVM attained best classification performance, followed by Logistic Regression & Naïve bayes, while least effectiveness is shown by Decision Tree. This study concludes that SVM is the most effective model for prediction of breast cancer among the six algorithms evaluated. Results show machine learning helps in timely & correct breast cancer detection & emphasizes choosing right models for reliable medical use.