Solving world hunger one graph at a time
May 18th, 2016
As a data nerd, it can be hard to imagine how our data and graphs can really help the world. Of course we are making people very happy with our stamps applications and great rewards, but we’re not exactly saving the world.
So after seeing a description of a new hackathon where we tried to help tackle world hunger on our internal Facebook group, it took very little convincing for us to sign up. Sure, it would cost us a day, but solving world hunger just seemed more important than that for some reason.
Arriving there on Saturday, it was quickly apparent that we were one of the few groups that signed up with our entire team. Sporting our fantastic onesies with some quotes from our founders, we were clearly winning the “best dressed” award this day.
After a lengthy explanation of the world of the World Food Programme and the issues they were dealing with, we were presented with two assignments:
1.Create a data model that helps predict where people might go hungry on a granular level
2.Create a dashboard on which managers are able to clearly see which regions are in need of food assistance the most
We opted to go for challenge two, mostly because out of the 4 of us, only 2 are true data modeling nerds, whereas the other two focus more on data visualization and presentation. In challenge two we were best able to combine our skills. Our goal was to create a single metric per region of Sierra Leone for managers to determine where help was most needed.
As usual at hackathons, we were too ambitious. We would have loved to incorporate some external data like weather forecasts in our visualization, but we soon noticed that time went a lot faster than we were producing results. We decided that we should limit ourselves to the data sets we were given, and just make a single awesome visualization in our tool of choice, Microsoft PowerBI.
From that moment on, everything went quite smoothly. Our data experts were combining data from multiple types of questionnaires, my colleague and me focused on determining the required data for the visualization and starting to clean the data where needed.
In the end, we were very happy with the final result. It looked nice, it was interactive, and it combined two sources of data and joined them on a district level. We didn’t win any awards, but we had a lot of fun, and I learned about new functionality in PowerBI that I wasn’t previously aware of.
Overall, the event was a success. Although we didn’t quite agree with the winner that seemed to be awarded for ambition, not for success, we walked away with a feeling that we could contribute, but also with a realization that solving world hunger is really hard.