Detecting Breast Cancer: A Comparative Study of Various Machine Learning Models
Pawan Patil, Bhushan Fulkar, Mansi Sharma, Rajendra M. Rewatkar · 2024
One of the deadliest illnesses in the world, cancer primarily affects women. Therefore, the primary goal of our research must be to cure cancer through scientific means. The secondary goal should be to detect cancer early on, as this can aid in the eventual removal of the cancer. After looking over 41 publications, we discovered that there are a number of methods for detecting cancer. In the medical field, detecting breast cancer is crucial since a correct diagnosis may greatly enhance patient outcomes. In order to assess how well different machine learning models identify breast cancer, we compare them all in this research. In particular, we contrast the Naive Bayes, Random Forest, Support Vector Machines (SVM), Logistic Regression, and Decision Tree methods. Our goal is to evaluate how well these algorithms classify cases of breast cancer.Of all the models studied, Random Forest performs better in terms of accuracy. Healthcare practitioners may find Random Forest to be a potential tool as it generates excellent accuracy rates in identifying breast cancer patients by utilizing its ensemble learning technique. Our comparative review provides useful details on the advantages and disadvantages of several machine learning methods for breast cancer. This research compares and contrasts several machine learning methods, including deep learning techniques, and finds that our suggested approach performs better than all of them.