Web based application for breast cancer detection using support vector machine

Due to the life-threatening nature of breast cancer, which predominantly affects women, early detection is critical for improving patient outcomes. Traditional screening methods, such as mammography, clinical examinations, and biopsies, are widely employed in the healthcare sector. However, these approaches face challenges, particularly in the misclassification of tumors. In this study, we developed a web application utilizing a Support Vector Machine (SVM) algorithm to create a predictive model for breast cancer that accurately distinguishes between benign and malignant tumors. Feature selection was employed to identify the most informative and relevant variables in the dataset, thereby mitigating the curse of dimensionality and enhancing model performance. To ensure accessibility, the predictive model was integrated into a web application, allowing medical professionals to use the tool for informed decisionmaking. Experiments were conducted using the publicly available Wisconsin Diagnostic Breast Cancer (WDBC) dataset, with results demonstrating a notable improvement in accuracy compared to similar studies.