Annelies Riezebos studies an MSc in Statistical Science for the Life and Behavioural Sciences, with a specialisation in Data Science. She works with us through a collaboration between 510 and Ortec, and is currently writing her thesis.
How did you hear about 510?
Someone from Ortec came to a presentation day at my university, where several organizations presented their thesis projects. This way, I got introduced to both the project and 510. Then realized I had heard of 510 before, as their work after hurricane Irma in St. Maarten had appeared on Dutch television.
I actually now think I should have started volunteering for 510 much earlier, as 510’s work really connects to my interests!
What project are you currently working on?
I’m working on impact data recognition using text mining techniques. In particular, I’m looking at documents related to the Disaster Relief Emergency Fund (DREF). This means that I use DREF-documents as a text case to see what these kinds of documents look like, and how impact data is described in them. Right now, I have a large set of labelled sentences. I want to create a model that can recognize impact data in these kinds of texts.
It was a new project, so I got to structure it myself and define my own research topic. Ortec and 510 had a link before, but this was the first project they did together.
510’s purpose is ‘Improve speed, quality and cost-effectiveness of humanitarian aid by using data & digital products.’ How are your skills helping 510 reach its purpose?
My work falls under the Predictive Impact Analytics category, and the Impact Based Foracting product. Within this category, there is a huge lack of historical impact data, which means that all sources of freely available text that might contain this data can help to improve our models. I’m helping to build a framework on how we can get good quality historical impact data from these sources automatically. This data can then be used in Forecast Based Financing.
The premise is that every dollar spent before a disaster hits, is better spent than one after. Taking that into account, I think I mostly help to improve both cost-effectiveness and quality of humanitarian aid using data.
My time at 510 was very inspiring: people see possibilities for improvement everywhere, and work hard to make this improvement happen as well. I hope I can stay as a volunteer, even after I start working at my new job at national government of The Netherlands. I’ll definitely take the experiences and the work ethic I learned at 510 with me!