Breast Tumor Prediction Using SVM with Rain Fall Optimisation Algorithm

Growth of breast cells in abnormal shape is termed as Breast Tumor (BT) over the past few years BT has become serious illness among women which has high death rates. According to the report, one in eight women worldwide will develop breast tumor at some point in their lives. There are other traditional methods which are time consuming as well as inefficient in diagnosing BT as well they are not effective in diagnosing BT at early stages which leads to detection of tumor when situation is worse and then there are chances that tumor have started spreading to other parts of the body; handling situation in worse condition is also critical for doctors so it is very important to diagnose the tumor at early stages. From the past few years, technology has changed to an extent that now computers are efficient in diagnosing the disease with help of Machine Learning (ML) algorithms. There are various models in ML like Decision Tree, KNN, Random Forest, and Linear Regression and many more but in comparison of accuracy of other models with SVM, SVM performs better. In this study, the author has proposed BT classification with use of Support Vector Machine (SVM) model which gives higher accuracy in comparison to other ML models. In order to improve the accuracy of SVM model the author have proposed the Rain Fall Optimisation Algorithm (RFOA), a meta-heuristic algorithm influenced by the natural world, tracks the motions as raindrops as they descend from the sky to the earth's surface which will improve BC classification accuracy by identifying the best SVM parameters. As per the experimental results, SVM with RFOA will improve the accuracy on Breast Cancer Wisconsin Original Data Set (BCWODS).