Potential Breast Cancer Drug Prediction using Machine Learning Models
N. Priya, G. Shobana · 2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE) · 2020
In recent years, researchers are working to produce best medication for Breast cancer, which has become a common disease among women. Several new drugs are synthesized and tested against the cell lines of the affected tissues. When different types of cell lines have to tested for its response to numerous drugs, the experimental costs are relatively high. Machine learning models helps in selecting highly potential and relevant drugs. Whenever a new compound is synthesized, it can be examined for its use as a breast cancer drug. Recent research classifies the cancer as benign or malignant type using machine learning techniques which employs numerical or medical images as input data. We propose the use of machine learning models in classifying the drug as a potent breast cancer drug. This paper investigates the application of feature reduction technique in further improving the prediction accuracy of the machine learning models. The performance of Logistic Regression, Support Vector Machine and Decision Tree models were evaluated and SVM provided optimal prediction accuracy in the process of breast cancer drug classification.