Analogizing of Evolutionary and Machine Learning Algorithms for Prognosis of Breast Cancer

摘要

Breast cancer has proven to be a serious disease caused in women according to medical science. This study focuses on prediction of breast cancer in three different datasets, namely: Wisconsin breast cancer (WBC), Wisconsin Diagnosis Breast Cancer (WDBC) and Wisconsin Prognosis Breast Cancer (WPBC) datasets. The comparative study has been done between evolutionary algorithms and machine learning algorithms. Evolutionary algorithms include Particle Swam Optimization (CPSO) and Genetic Algorithm for Neural Network (GANN) whereas machine learning algorithms include KNN and C4.5 for predicting the breast cancer. The results are obtained after performing the experiment on different algorithms on the basis of their accuracy and standard deviation which may help people in medical science for better prediction of their disease and hence enabling appropriate treatment.