Comparative Analysis and Visualization of Breast Cancer using Machine Learning Models
Akanksha Rani, Nonita Sharma · 2022 10th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO) · 2022
In recent decades, humans especially women are now commonly threatened by breast cancer, which has a high morbidity and fatality rate. It is challenging for doctors to develop a consistent treatment solution because there aren't any reliable prognostic models available. Therefore, it takes time to establish a method that produces the least amount of error in order to improve accuracy. To predict the outcome of breast cancer using various datasets, this publication analyses and contrasts four state-of-the-art models: Support Vector Machine (SVM), Logistic Regression, Linear Discriminant Analysis (LDA), and K-Nearest Neighbor (KNN). On the Google Collaborator platform, all experiments are carried out in a simulation setting. The research's objective is divided into three categories. Prediction of cancer before a diagnosis is the first domain, while The second domain focuses on forecasting diagnosis and therapy, while the third domain is concerned with how treatments will turn out. The proposed work can be utilized to forecast the results of various procedures, and based on the need, relevant techniques can be applied. This study compares and assesses how well the models anticipate the future. Future studies can be conducted to forecast the other various criteria, and research on breast cancer can be grouped according to these other parameters.