A comparative study on predicting influenza outbreaks using different feature spaces: application of influenza-like illness data from Early Warning Alert and Response System in Syria

Ali Darwish,Assef Jafar,Yasser Rahhal

doi:10.1186/s13104-020-4889-5

Ali Darwish, Assef Jafar + Show 1 more

Open Access

https://doi.org/10.1186/s13104-020-4889-5

Copy DOI

Abstract

ObjectiveAn accurate forecasting of outbreaks of influenza-like illness (ILI) could support public health officials to suggest public health actions earlier. We investigated the performance of three different feature spaces in different models to forecast the weekly ILI rate in Syria using EWARS data from World Health Organization (WHO). Time series feature space was first used and we applied the seven models which are Naïve, Average, Seasonal naïve, drift, dynamic harmonic regression (Dhr), seasonal and trend decomposition using loess (STL) and TBATS. The Second feature space is like some state-of-the-art, which we named 53-weeks-before_52-first-order-difference feature space. The third one, we proposed and named n-years-before_m-weeks-around (YnWm) feature space. Machine learning (ML) and deep learning (DL) model were applied to the second and third feature spaces (generalized linear model (GLM), support vector regression (SVR), gradient boosting (GB), random forest (RF) and long short term memory (LSTM)).ResultsIt was indicated that the LSTM model of four layers with 1-year-before_4-weeks-around feature space gave more accurate results than other models and reached the lowest MAPE of 3.52% and the lowest RMSE of 0.01662. I hope that this modelling methodology can be applied in other countries and therefore help prevent and control influenza worldwide.

Highlights

It was indicated that the Long short term memory (LSTM) model of four layers with 1 − year − before_4 − weeks − around feature space gave more accurate results than other models and reached the lowest mean absolute percentage error (MAPE) of 3.52% and the lowest root mean squared error (RMSE) of 0.01662
Some researches treated the problem as an instance of more general time series forecasting using time series methods (ARIMA, ARIMA-seasonal and trend decomposition using loess (STL), GARMA) [9, 10, 17, 27], while others used Machine learning (ML) methods including Stacked linear regression [24, 26], AdaBoost regression with decision trees [26], : gradient boosting (GB) [12], : support vector regression (SVR) [26, 28], elastic net
We proposed novel future spaces, namely n − years − before_m − weeks − around, and compared to some existing future spaces that utilize historical observations in different ways by integrating it into state-of-the-art ML and deep learning (DL) models

Summary

Results

It was indicated that the LSTM model of four layers with 1 − year − before_4 − weeks − around feature space gave more accurate results than other models and reached the lowest MAPE of 3.52% and the lowest RMSE of 0.01662.

Introduction

Main text

Discussion