Abstract: The machine learning methodology consists of two stages: the training stage, during which an algorithmic classification program is taught to assign labels to data, and the testing stage, during which the algorithmic classification program is put through its paces. Data classification, also known as supervised learning, is a type of data processing in which data is divided into predetermined categories.
Two major ensemble machine learning classifiers, ReLU and Sigmoid, are the focus of our investigation. ELM is a novel approach to regression and classification issues that originates from single-hidden-layer feedforward neural networks. Classifiers such as ELM, SVM, KNN, Logistic regression, and others will be used in this work to analyze medical datasets with the help of activation functions. Diabetes, lung cancer, and brain tumor data sets are included. For some activation functions, including ReLU and sigmoid, ELM is optimized for classification performance. Using a publicly available, free online dataset of patients with mild to severe disease difficulties, we will employ ELM's Sigmoid, Relu, and other classifiers to analyze data on 900 to 1000 patients in the medical field.
Keywords: Extreme Learning Machine, Sigmoid, ReLU, SVM, KNN, J48, Logistic Regression, Artificial Intelligence, Machine Learning, and Deep Learning.
Title: The Performance Comparison Of Ensemble Machine Learning Classifiers On Medical Datasets
Author: Mohammed Lawan, Prof. Engin Avci, Prof. Dr. Beşir Dandil, Erhan Turan
International Journal of Engineering Research and Reviews
ISSN 2348-697X (Online)
Vol. 11, Issue 1, January 2023 - March 2023
Page No: 27-35
Research Publish Journals
Website: www.researchpublish.com
Published Date: 08-March-2023
