An Approach to Predict Breast Cancer Using Ensemble Learning Technique

Breast cancer stands as the second most prominent contributor to cancer-related fatalities and ranks as the most prevalent form of cancer diagnosed among women worldwide. Although certain devices are designed for breast cancer detection, they frequently produce false positive outcomes, compelling patients to endure unnecessary and costly surgical procedures, often accompanied by discomfort. In this paper ensemble learning-based technique to predict the occurrence of breast cancer is presented. The ensemble learning technique refers to the use of multiple machine learning algorithms that aggregate the results of numerous models to solve classification and/or regression problems. It aims at improving predictability in models by combining several models to make one very reliable model. The different machine learning algorithms utilized by the proposed scheme include Support Vector Machine (SVM), Naïve Bayes (NB), K-Nearest Neighbors (KNN) and Random Forest (RF). The method proposed in this paper utilizes the most prominent features from the dataset and predicts the occurrence of breast cancer accurately. The results of the proposed model indicate that the ensemble voting approach is ideal for breast cancer prediction as it gives better accuracy.