UNICEF Arm 2030 Vision #1: Flood Prediction in Malawi
$10,000 USD
Predict flood extent caused by storms in southern Malawi
1596 data scientists enrolled, 480 on the leaderboard
2 December 2019—17 May 2020
Bad data, I'm disappointed
published 28 Mar 2020, 18:33

I think rain data is bad for the purpose of the competition. These data is given in very large homogeneous squares and this means that rainfall hardly influences the prediction. Does anyone think something like this?

Yeah, the most predictive feature is the elevation :)

take your time to explore The Elevation; find the slope and more feature engineering... thats where the power of the predictor lies...

Night of codes. Can we chat via Linkedin:

You can find me here to connection on Linkedin. Thanks.


Haha I feel you I'm a civil engineer myself and the data is really lacking. Finding the data on it's own is the hardest part as opposed to actually analysing it