Project Benson Begins
Project Benson has begun.
Wow. We hit the ground running alright.
First day of the bootcamp and we’re busy on the first project.
The projets throughout the Metis Bootcamp seem to trend in a couple of ways. One trend is from a group effort to a solo effort. For this first project, my partners in crime are Dean, Derrick and Kendall. We sequestered ourselves at the sunny side of the floorspace and barricaded ourselves from the rest of the world as we discussed how best to attack the problem.
Kendall and I were the first to find the data. But we found it in different places. And it’s good that we did so. The MTA site provides the data in smaller chunks, which should prove more amenable to rapid prototyping. But as Kendall dug into the data available on the New York site site, we realized it was the same data, same columns, but… the data was available in entirety, millions of rows of it.
We chatted about what to look for… or how to use the data available. Did we care about location down the level of which street exit of any particular station? Should we focus on Spring traffic since the problem statement suggested an event at the start of Summer? Dare we try to accommodate or consider the effect of Spring Break?
We looked at the data enough to realize we had some data munging to do so we clearly needed something in the fashion of people per unit time at any station while the turnstile data is just reporting a (supposedly) ever increasing count of the actual turnstile reader.
We broke with the idea that we should first play with the data more and then we could scale or filter as we chose. Furthermore, Dean stated he would attempt to work in Tableau since he had access to that. This is one of those benefits to the group aspect of the Metis Bootcamp. We all bring different things to the table and can share and learn from one another.