Open AccessThe Comparative Study for Predicting Disease Outbreak
Author(s) -
Anifatul Faricha,
Mehak Nanda,
Siti Maghfirotul Ulyah,
Ni'matut Tamimah,
Enny Indasyah,
Robin Addwiyansyah Alvaro Samrat
Publication year - 2020
Publication title -
journal of computer, electronic, and telecommunication
Language(s) - English
Resource type - Journals
eISSN - 2723-5912
pISSN - 2723-4371
DOI - 10.52435/complete.v1i1.48
Subject(s) - support vector machine , kriging , mean squared error , process (computing) , statistics , computer science , regression , data mining , cross validation , artificial intelligence , machine learning , pattern recognition (psychology) , mathematics , operating system
To know the prediction of disease outbreak, proper predictive modeling is required to represent the dataset. This study presents the comparative predictive modeling for predicting disease outbreak using two models i.e., optimizable support vector machine (SVM) and optimizable gaussian process regression (GPR). The dataset used in this study contains three cases i.e., positive cases, recovered cases, and death cases. The dataset at each case is divided into training dataset for the training process and external validation dataset for the validation process. Based on the training process and validation process, the root mean square error (RMSE) at positive cases, recovered cases, and death cases using optimizable GPR is substantially more effective for prediction than the optimizable SVM. According to the result performance, by applying optimizable GPR, the training process has the average RMSE of 19.54 and the validation process has the average RMSE of 15.85.