Open AccessDengue Outbreak Prediction Using an Improved Salp Swarm Algorithm
Author(s) -
Khairunnisa Amalina Mohd Rosli,
Zuriani Mustaffa,
Yuhanis Yusof,
Mohamad Farhan Mohamad Mohsin
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/769/1/012031
Subject(s) - dengue fever , mean squared error , outbreak , mean absolute error , algorithm , statistics , artificial intelligence , computer science , geography , machine learning , data mining , mathematics , medicine , virology
Dengue disease is the most common type of disease caused by mosquitoes. It is reported that dengue fever was first recognized in Thailand and Philippines in 1950. According to World Health Organization (WHO), dengue is a viral disease that spread in public environment where the number of dengue cases reported enlarge within 5 years by 1 million from 2.2 million in 2010 to 3.2 million in 2015. Until today, numerous studies by researchers to improve the prediction of dengue fever disease based on Computational Intelligence (CI) methods have been reported. The research includes study using Swarm Intelligence (SI) algorithm. This research study proposed an improved Salp Swarm Algorithm (iSSA) for dengue outbreak prediction. The original SSA will be enhanced by enriching the exploration and exploitation process for the sake of improving the accuracy of dengue outbreak prediction. This will be done by inducing a mutation based on Levy Flight. Later, the iSSA algorithm will be realized on dengue disease dataset. The proposed iSSA will be compared against the original SSA and another CI method known as Grey Wolf Optimization (GWO). Two evaluation indicators known Root Mean Square Error (RMSE) and as Mean Absolute Error (MAE) are proposed in this research study to evaluate the prediction model where the smaller the value obtained, more accurate the prediction model. The result demonstrated that the proposed model produces a better result compare two the other results where the value of MAE and RMSE of the proposed model is smaller compare to other two model.