An LSTM approach to predict migration based on Google Trends

Details

Supervisors

Schaus, Pierre ; Kiossou, Harold Silvère

Faculty

Ecole polytechnique de Louvain
Ecole polytechnique de Louvain

Degree label

Master [60] en sciences informatiques
Master [120] : ingénieur civil en informatique

Abstract

Being able to model and predict international migration as precisely as possible is crucial for policy making. Recently Google Trends data in addition to other economic and demographic data have been shown to improve the prediction quality of a gravity linear model for the one-year ahead predictions. In this work, we replace the linear model with a long short-term memory (LSTM) approach and compare it with two existing approaches: the linear gravity model and an artificial neural network (ANN) model. Our LSTM approach combined with Google Trends data outperforms both these models on various metrics in the task of predicting the one-year ahead incoming international migration to 35 Organisation for Economic Co-operation and Development (OECD) countries: for example the root mean square error (RMSE) and the mean average error (MAE) have been divided by 5 and 4 on the test set. This positive result demonstrates that machine learning techniques constitute a serious alternative over traditional approaches for studying migration mechanisms.