A Tale of Two Hackathons

I just thought I would write a quick note about two hackathons I participated in recently.

About two weeks ago I went to a hackathon hosted by BetaNYC, an organization looking to better New York City through data. It was a great day of work learning about Civic Hacking through improved Citygram notifications. I was surprised by the feeling of accomplishment once 7:00pm arrived and work ceased, despite none of my notifications being pushed through. The project is definitely one I’ll continue my participation in. At the moment I have a notification on school’s with maintenance issues in the pipeline ready to be integrated into the service.

The second hackathon I was involved in was Get Better With Data. Over Friday night and most of Saturday, people gathered at the offices of Insight Data Science to work with health data. Out of the three tracks, I was placed in modeling (Yes!), and my teammates and I spent Saturday doing just that.

It was harder than I thought it would be. The Chronic Disease Health data was a mess, and just trying to wrap my head around left me nursing a headache on my way home. In the end, we were able to build several models to extract some meaning. The model I was most involved in involved using Random Forests to get a sense of the relative importance of each chronic diseases to the average payment made for each Medicare plan.

I didn’t get around to doing more advanced modeling techniques, but the results seemed pretty good. I came to that conclusion after a conversation with Brendan Stern of Dataiku shared with his knowledge of public health with me, and also finally made me understand the importance of developing a story after all the analysis has been said and done. Here’s the notebook of my work.

Both events were fulfilling experiences I would like to have again.

Written on November 8, 2015