Of the many causes of deaths among women in the world, breast cancer is ranked highly with treatment outcomes largely dependent on early diagnosis. In this paper, performance of various machine learning algorithms applied to distinguishing between benign and malignant cases of breast cancer is examined considering Breast Cancer Wisconsin (Diagnostic) Dataset availed by Kaggle. Some aspects were extracted based on images of digitized fine needle aspirate (FNA) of breast masses, with an image holding features on cell nuclei. Preprocessing of the dataset included the removal of irrelevant columns, encoding target labels and feature scaling. Four machine learning models namely LightGBM, CatBoost, Multi-Layer Perceptron (MLP) and Quadratic Discriminant Analysis (QDA) as well as an ensemble method were tested. The evaluation scores were Accuracy, AUC-ROC and Average Precision. The MLP model attained the maximum accuracy (0.9825) and the high AUC-ROC (0.9981) was recorded by the ensemble model. Results indicate that deep learning based methods and ensemble methods offer better predictive scores. This paper shows the possibility of using multi-machine learning approaches to detect breast cancer in a long-lasting and consistent way.