v1.1.1

Python code (scripts, libraries, and Jupyter notebooks) used to generate the camera ready version of the paper "A Comparative Analysis of Wealth Index Predictions in Africa between three Multi-Source Inference Models" paper published in the proceedings of the ECMLPKDD 2024 conference and presented at the SoGood'24 workshop (latest preprint: https://bit.ly/PovertyMapsECMLPKDD24-preprint)

Full Changelog: ecmlpkdd24-Aug2024...ecmlpkdd24-CR

v1.1.0

Python code (scripts, libraries, and Jupyter notebooks) used to generate results for the "A Comparative Analysis of Wealth Index Predictions in Africa between three Multi-Source Inference Models" paper published in the proceedings of the ECMLPKDD 2024 conference (preprint: https://arxiv.org/abs/2408.01631)

v1.1.0:

• Updates the code from the WWW23 paper (scripts and libs)
• Adds the comparison analysis and results (including new predictions) of 6 African countries (SLE, LBR, UGA, RWA, ZAF, GAB)

What's Changed

• Comparison between RWI and IWI predictions by @lisette-espin in #1

New Contributors

• @lisette-espin made their first contribution in #1

Full Changelog: https://github.com/lisette-espin/PovertyMaps/commits/ecmlpkdd24-Aug2024

v1.0.1

Python code (scripts, libraries, and Jupyter notebooks) used to generate results for the "Interpreting wealth distribution via poverty map inference using multimodal data" paper published at The Web Conference 2023.

v1.0.1: calculates the mean pearson corr to generate Figure 3 in paper (instead of the pearson corr of all runs)

v1.0.0

Python code (scripts, libraries, and Jupyter notebooks) used to generate results for the "Interpreting wealth distribution via poverty map inference using multimodal data" paper published at The Web Conference 2023.