Targeting disaster relief from space

For my final project at Metis, I used machine learning to better target disaster relief efforts. I focused on Typhoon Haiyan, which hit the Philippines in November of 2013. It broke records for having the highest wind speeds upon landfall and destroyed over 1 million homes.

After natural disasters, it’s important to understand which areas suffered the most damage in order to prioritize relief efforts. Often times damage assessment maps are created by volunteers with the Humanitarian Open Street Map team who compare satellite imagery before and after the disaster and manually label each building with their evaluation of damage. However these maps are time and labor intensive to create, and not always accurate.

Continue reading “Targeting disaster relief from space”

The difficulties of drought data

Over the last few weeks, I’ve been working with monthly rainfall and temperature data in an effort to predict drought in India on a district level. My initial hypothesis was: since climate change entails more extreme temperatures and more extreme weather events such as drought, can I use temperature data to predict drought one year ahead?

This data was challenging as it was time series data with multiple, imbalanced classes. In this post, I want to talk about some lessons learned from working with this difficult data.

Continue reading “The difficulties of drought data”