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”
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”
For my second project at Metis, I scraped a beer review website and utilized supervised learning models to predict beer ratings with a low complexity, high R-squared random forest regressor.
Continue reading “Feature engineering + supervised learning”
Week two felt far easier than week one of Metis. The first week is understandably overwhelming. They say it’s like drinking from a fire hose. And I spent most of my energy trying not to drown — trying to make headway on a group project with an unreasonably tight deadline, attempting to absorb all the new information being thrown at me left and right in lecture, and working to complete my individual challenges. Add in navigating a new commute, meeting new people, and communicating in what is essentially a new language. Needless to say, it was a lot.
Why did week two feel easier? I didn’t automatically become a sponge or python guru overnight nor did I learn how to add hours to my day.
Continue reading ““Hey, can you help me with this?””
I’ve recently left my research job to pursue a full time data science bootcamp at Metis. Some people have commented to me that this seems like a shift as I’ve worked in international development research the last five years. And they’re right that I’ve been reading and writing, not coding in Python. But that doesn’t mean data science is an entirely separate undertaking. To me, data science continues many of the threads that have been present in my work and life for years.
Continue reading “Why data science?”
Twelve women stand in a row, ankle-deep in an irrigated field, submerging rice seedlings as quickly as they can. The work is meticulous. Paddy fields stretch for miles, broken up by palm trees and mango groves. Monsoons are coming soon, the farmers say. And hopes are high the rains will mean much better harvests than the droughts of the last two years.
I’m looking on from the side of a road in rural India in 100 degree heat — a senior research assistant 9,000 miles from my office at the Stanford Center for International Development — trying to find answers to seemingly intractable questions: Despite this promising expanse of newly planted fields, why are so many farmers trapped in debt? And what can be done about it?
Continue reading “Looking at rural debt through the eyes of India’s farmers”