This is the backyard of Badass Datenjungs (distracted franklin), herzlich willkommen!
In the beginning we wanted to use crop yield data from kaggle and satelite images from google earth engine to anaylze and predict future crop yield. Unfortunately the kaggle data was of low quality and many other teams predicted crop yield from satellite data before us. We now settled on predicting prevalence of undernourishment in a given coutry and year. We use climate data (earth engine), economic data (FAO) and violent conflict data (UCDP) to do so. Feel free to have a look around :)
Team members: Neelesh - https://github.com/neelblabla ; Nils - https://github.com/nilsmart96 ; Jerome - https://github.com/sonic171717
Description of files:
-
Earth Engine Data Collection -> As the name suggests, python scripts that collect data from google earth engine
-
Dataset Construction -> Steps that are needed to build our final dataset, combining climate, economics and conflict data
-
Models -> Analysing our self-created dataset and making predictions
-
Data -> Raw input datasets as well as everything that is being created during the project
This study is now available as a micropublication on the following link: https://data.flashpub.io/articles/kacqxqvlqfh72mxyb6nrjkjyocskwvoagd2gxbsiutvl2pqepgczvhq
Further information can be found here: https://docs.google.com/document/d/1LRx-RHb_BrDnaqMHl_3nnihoNGNaeNGI/edit#heading=h.30j0zll