Automatic detection and prevention of breast cancer by ML algorithms with dimensionality reduction technique PCA

Early detection and prevention of breast cancer are vital for improving patient outcomes and reducing mortality rates. Machine learning algorithms, coupled with principal component analysis (PCA), offer a promising approach for enhancing the accuracy and efficiency of breast cancer diagnosis. This paper proposes a novel approach to early breast cancer detection, combining machine learning algorithms, including support vector machines, neural networks, and random forests, with PCA. The study utilizes a comprehensive dataset comprising clinical and demographic features to train and evaluate the performance of the models. Results demonstrate the efficacy of the proposed approach in accurately identifying breast cancer at its early stages, thus enabling timely intervention and improved patient care. Furthermore, the integration of PCA aids in reducing dimensionality and enhancing the interpretability of the models thereby is facilitating more effective decision-making in clinical settings. Overall, this research underscores the potential of machine learning algorithms with PCA in revolutionizing breast cancer detection and prevention strategies, ultimately contributing to better healthcare outcomes and patient wellbeing. In this study we have applied five ML technique: Logistic Regression, Decision tree, Random forest, Naïve byes and support vector machine on the Breast Cancer Wisconsin Diagnostic dataset. The evaluation of classification outcomes was conducted both with and without the implementation of PCA. Notably, upon applying PCA, the random forest classifier emerged as the top performer, surpassing all other classifiers. Impressively, it attained an accuracy rate of 99.8%, showcasing its superior performance in the dataset.