Breast cancer is an undeniably fatal ailment affecting women and is a significant global concern. Moreover, with its rampantly increasing death rates, detection in early stages would help combat the disease. Machine learning (ML) proves to be a coherent solution for its detection, and in this paper, we have implemented four ML models on Wisconsin Diagnostic Breast Cancer (WDBC) dataset, namely support vector machine (SVM), gradient boosting (GB), logistic regression (LR), and K-nearest neighbours (KNN), for the prediction. The implementation of these models has been coupled with hyperparameter tuning to get the best estimate of each model’s performance. The results are observed after inclusion of Neighbourhood Component Analysis (NCA) for dimensionality reduction in the dataset, to observe how models perform when paired with it. Our results show that KNN had the highest accuracy and thus, the best performance.