Open AccessComparisons Among Multiple Machine Learning Based Classifiers for Breast Cancer Risk Stratification Using Electrical Impedance Spectroscopy
Author(s) -
Md. Toukir Ahmed,
Md. Rayhanul Masud,
Abdullah Al Mamun
Publication year - 2020
Publication title -
european journal of electrical engineering and computer science
Language(s) - English
Resource type - Journals
ISSN - 2736-5751
DOI - 10.24018/ejece.2020.4.4.227
Subject(s) - random forest , c4.5 algorithm , machine learning , artificial intelligence , naive bayes classifier , computer science , decision tree , multilayer perceptron , perceptron , breast cancer , classifier (uml) , support vector machine , artificial neural network , medicine , cancer
Nowadays, women worldwide are affected greatly by Breast cancer. The consequences of the disease and the number of affected are so alarming that it requires immediate attention. Prediction of the disease is the primary and most significant route to prevention of it. This study aims to have a comparison among multiple machine learning based classifiers for breast cancer risk stratification using resonance-frequency electrical impedance spectroscopy. Five machine learning based classifiers namely- Naïve Bayes, Multilayer perceptron, J48, Bagging and Random Forest were applied to the dataset and a comparison was made based on different performance metrics. The study demonstrated that Random Forest classifier performed slightly better than the others in both splitting and folding of the dataset.