Breast Cancer Prediction Using Nature Inspired Algorithm

Abstract

Medical industry, though various researches over decades, has figured out breast cancer to be one of the most common diseases in women. Studies have shown that every eighth woman is suffering from it. This research has been done with the intent to predict the occurrence of breast cancer with the help of various machine learning algorithms. For the analysis purpose, three different datasets were utilized, Wisconsin Breast Cancer (WBC) dataset, Wisconsin Prognosis Breast Cancer (WPBC), and Wisconsin Diagnosis Breast Cancer (WDBC) dataset. Also, the classification used for these datasets was done using Hierarchical Decision Tree (HIDER), PSO, and Genetic Algorithm for Neural Network (GANN). The results were compared based on the accuracy achieved was found that HIDER showed the best results with WBC Dataset, while GANN was the most accurate one with WDBC and WPBC datasets. This research would help organizations, working in the health sector, especially in cancer studies, to predict breast cancer accuracy with accuracy and help cure it in the early stages.