ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

ON COMPARING MULTI-LAYER PERCEPTRON AND LOGISTICREGRESSION FOR CLASSIFICATION OF DIABETIC PATIENTS INFEDERAL MEDICAL CENTER YOLA, ADAMAWA STATE

Journal: INTERNATIONAL JOURNAL OF ENGINEERING TECHNOLOGIES AND MANAGEMENT RESEARCH (Vol.8, No. 6)

Publication Date:

Authors : ; ;

Page : 24-40

Keywords : Logistic Regression; MultiLayer Perceptron; Artiβicial Neural Network; Log Likelihood Ratio; Diabetes Mellitus;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

The logistic regression (LR) and Multi-Layer Perceptron (MLP) are used to han-dle regression analysis when the dependent response variable is categorical.Therefore, this study assesses the performance of LR and MLP in terms of clas-siβication of object/observations into identiβied component/groups. A data setconsistsof553casesofdiabetesmellituswerecollectedatFederalMedicalCen-ter, Yola. The variables measured: Age(years), Mass of a patient(kg/meters),glucose level (plasma glucose concentration, a 2-hour in an oral glucose toler-ance test), pressure (Diastolic blood pressure mmHg), insulin (2-hour seruminsulin mu U/ml) and class variable (0 or 1) treating 0 as false or negative and1 treated as true or positive test for diabetes. The method used in the studyis Logistic regression analysis and the multi-Layer perceptron, a type of Arti-βicial Neural Network, confusion matrix, classiβication, network algorithm andSPSS version 21 for Windows 10.1. The result of the study showed that LP clas-siβies diabetic patients correctly with 91.8% accuracy. While it classiβies non-diabetic patients with 89.1% accuracy. MLP classiβies diabetic patients with88.6% accuracy while it classiβies non-diabetic patients with 93.2% classiβica-tion accuracy. Overall, MLP classiβies better with 91% accuracy while LR clas-siβies with 90.6% accuracy. This study complements other literatures whereMLP, a type Artiβicial neural network classiβies and predicts better than othernon-neural network classiβiers.

Last modified: 2021-11-13 16:58:58