An LSTM approach to Predict Migration based on Google Trends

Codes by Nicolas Golenvaux & Pablo Gonzalez Alvarez.

This repository contains the companion codes for the paper An LSTM approach to Predict Migration based on Google Trends by Nicolas Golenvaux, Pablo Gonzalez Alvarez, Harold Silvère Kiossou, Pierre Schaus.

This paper is part of our joint master's thesis work.

Install requirements

pip install -r requirements.txt

Structure

Files without a comment have a descriptive name to understand what they contain:

/                                       I am (G)root!
    LICENSE.md                          MIT License
    README.md                           You are here
    extract_gti.py
    migration-gti-lstm.ipynb            Notebook to build models
    requirements.txt
    data/
        LICENSE.md                      CC-BY-4.0 License
        basicDataLSTM.pkl
        bestResultLSTM.pkl
        dataANN.pkl
        destination_countries.csv
        keywords.csv
        logDataLR.pkl
        origin_countries.csv
        replication_bilateral.dta
        resultANN_test.pkl
        resultANN_train.pkl

License

Except stated otherwise (cf. Credits), the content of this repository itself as well as the data are licensed under the Creative Commons Attribution 4.0 International License (CC-BY-4.0), and the source codes are licensed under the MIT License.

Credits

Data sources:
    Google Trends (https://www.google.com/trends)
    M. H. Böhme, A. Gröger, and T. Stöhr, "Searching for a better life: Predicting international migration with online search keywords" - Researcher Data (https://www.sciencedirect.com/science/article/pii/S0304387819304900#mmc2)
    OECD International Migration Database (https://www.oecd-ilibrary.org/content/data/data-00342-en)
    United Nations Department of Economic and Social Affairs Population Division UN DESA. International Migrant Stock 2019 (https://www.un.org/en/development/desa/population/index.asp)
    World Bank 2020 World Development Indicators (https://datacatalog.worldbank.org/dataset/world-development-indicators)

Software licenses:
    Keras (https://keras.io) - MIT License
    Matplotlib (https://matplotlib.org/) - Matplotlib License
    NumPy (https://numpy.org/) - BSD-3-Clause License
    Pandas (https://pandas.pydata.org/) - BSD-3-Clause License
    pytrends (https://github.com/GeneralMills/pytrends) -  Apache-2.0 License
    Scikit-learn (https://scikit-learn.org/) - BSD-3-Clause License
    Seaborn (https://seaborn.pydata.org) - BSD-3-Clause License
    Tensorflow (https://www.tensorflow.org) - Apache License 2.0