Predictive Analysis of Dengue Outbreak Based on an Improved Salp Swarm Algorithm

Abstract

The purpose of this study is to enhance the exploration capability of conventional Salp Swarm Algorithm (SSA) with the inducing of Levy Flight. With such modification, it will assist the SSA from trapping in local optimum. The proposed approach, which is later known as an improved SSA (iSSA) is employed in monthly dengue outbreak prediction. For that matter, monthly dataset of rainfall, humidity, temperature and number of dengue cases were employed, which render prediction information. The efficiency of the proposed algorithm is evaluated using Root Mean Square Error (RMSE), and compared against the conventional SSA and Ant Colony Optimization (ACO). The obtained results suggested that the iSSA was not only able to produce lower RMSE, but also capable to converge faster at lower rate as well.

Keywords

• dengue outbreak prediction
• meta-heuristic
• optimization
• predictive analysis
• salp swarm algorithm
• swarm intelligence