The Classification Performance using Support Vector Machine for Endemic Dengue Cases

Dengue fever (DF) and the potentially fatal dengue haemorrhagic fever (DHF) are continue to be a crucial public health concern in Malaysia. This paper proposes a prediction model that incorporates Support Vector Machine (SVM) in predicting future dengue outbreak. Datasets used in the undertaken stud...

Full description

Bibliographic Details
Published in:Journal of Physics: Conference Series
Main Author: Nordin N.I.; Mohd Sobri N.; Ismail N.A.; Zulkifli S.N.; Abd Razak N.F.; Mahmud M.
Format: Conference paper
Language:English
Published: Institute of Physics Publishing 2020
Online Access:https://www.scopus.com/inward/record.uri?eid=2-s2.0-85086720184&doi=10.1088%2f1742-6596%2f1496%2f1%2f012006&partnerID=40&md5=e4e3dce30ec143875accc6a75b7d582c
Description
Summary:Dengue fever (DF) and the potentially fatal dengue haemorrhagic fever (DHF) are continue to be a crucial public health concern in Malaysia. This paper proposes a prediction model that incorporates Support Vector Machine (SVM) in predicting future dengue outbreak. Datasets used in the undertaken study includes data on dengue cases provided by the Health Department in Kelantan, Malaysia. Data scaling were applied to normalize the range of features before being fed into the training model. In this regard, SVM models built on the basis of three different kernel functions including Gaussian radial basis function (RBF), polynomial function and linear function. The SVM with RBF kernel function was superior to the other techniques because it obtains the highest prediction accuracy of 85%. The polynomial is an alternative model that can achieve a high prediction performance in terms of sensitivity (76%) and specificity (87%). © 2020 Published under licence by IOP Publishing Ltd.
ISSN:17426588
DOI:10.1088/1742-6596/1496/1/012006