Hybrid Machine Learning Models for Improved Breast Cancer Prediction: A Comparative Study

Recently, breast cancer has become a major health issue for women. It is the second most common cause of death for women. Breast cancer is one of the most common types of cancer, second only to lung cancer. Age, gender, genetics, hormones, breast mass, and being overweight are all important factors that can lead to breast cancer. It is the most common disease in women and is responsible for a lot of deaths around the world. Many people who get breast cancer don't have a family history of it, but people who do have a family history are more likely to get it themselves. The chance of getting breast cancer goes up because of things like age, genetics, breast density, obesity, and radiation exposure. To tell the difference between benign and malignant tumors, it is very important to use accurate testing methods. Early detection and treatment of breast cancer are very important. For finding breast cancer, predictive methods, especially Machine Learning (ML) methods, are very important. The Wisconsin Dataset of Breast Cancer (WDBC) is used in the study to compare different machine learning methods. We look at five supervised machine learning classifiers: Support Vector Machine (10 features), K-Nearest Neighbor, Random Forest, and Naïve Bayes. To make a mixed model for predicting breast cancer, an ensemble learning method is used. Other models are less accurate than the Support Vector Machine (SVM) model, which gets 96.25% right. The suggested hybrid model can accurately predict 97.08% of cases of breast cancer.