Implementation Of Climate Factor Prediction Of Dengue Outbreak With Mosquito Spreading Control Using Optimal Feature Association Of Attention-Based Deep Learning Network

Abstract

Dengue fever and other mosquito-borne illnesses spread quickly and this disease has intricate connections between social, economic, and ecological factors, but weather patterns including variations in temperature, humidity, and rainfall continue to have a significant impact. Therefore, an automated mosquito spreading control system is implemented using climatic data and a deep learning-based dengue outbreak prediction framework. The climate factors data for dengue fever outbreak is predicted in this work for controlling the mosquito spreading. Initially, the requisite climate data such as temperature, humidity, precipitation, and so on is gathered from the benchmark resources and the collected data is fed into the optimal feature selection stage. Here, the necessary feature of the mosquito spreading control process is optimally selected by utilizing the Modified Golden Eagle Optimizer (MGEO). Further, the optimally selected features are forwarded to the Attention-based Temporal Convolutional Network (TCN) with Long Short Term Memory layer (ATCN-LSTM) for predicting the dengue outbreak. Subsequently, the climate factors data for dengue fever outbreak is predicted by analyzing and varying the climatic parameters and its micro-level.  Here, the parameter and its micro-level are determined optimally by the same MGEO algorithm to control mosquito spreading.