Breast Cancer Wisconsin Diagnosis using KNN and Cross Validation
Prof. Mrs. Neha Singh, Ayush Yadav, Nitesh Sarkar · International Journal of Research Publication and Reviews · 2025
Breast cancer remains one of the most prevalent malignancies affecting women globally, underscoring the urgent need for effective diagnostic methodologies.This study examines the application of logistic regression as a predictive modeling technique for diagnosing breast cancer, utilizing the Wisconsin Breast Cancer Dataset available on Kaggle.The dataset encompasses critical features derived from digitized images of fine needle aspirate (FNA) samples, including attributes related to cellular size, shape, and texture.Our methodology involved a comprehensive preprocessing phase, incorporating normalization and the treatment of missing values, followed by a rigorous training and validation process employing k-fold cross-validation to ensure the robustness of the model.The logistic regression model achieved notable performance metrics, including an accuracy rate of X%, alongside sensitivity and specificity values of Y% and Z%, respectively, thus indicating its potential utility in clinical diagnostics for early breast cancer detection.Furthermore, we analyzed the model's coefficients to identify significant predictors of malignancy, thereby enhancing the understanding of the underlying factors associated with breast cancer.This research underscores the efficacy of logistic regression as a straightforward yet potent tool for diagnostic applications, setting the stage for further exploration of advanced machine learning techniques in oncological prognosis.