For my final project at Metis, I wanted to choose something that enabled me to incorporate all that I had learned during the past three months. Predicting Portland home prices allowed me to do this because I was able to incorporate various web scraping techniques, natural language processing on text, deep learning models on images, and gradient boosting into tackling the problem.

There has been much talk over how much debt students are saddled with as they graduate from college. The most aggregious cases involving for-profit colleges such as University of Phoenix, DeVry, ITT, and Everest College have been well-documented, including on John Oliver and in government reports. However, even programs as highly regarded as Harvard’s drama program and Johns Hopkin’s music program have received failing Debt-To-Earning ratings by the government, as chronicled by the New York Times.

Our second project at Metis involved using scikit-learn and regression models in order to analyze a data set of our choosing. I chose to analyze fantasy football performance, of course, as I wanted to improve my chances of slaughtering my fantasy opponents this season. Each pre-season, ESPN puts out projected rankings of how each player will perform. In September of 2016, ESPN released the top 300 player rankings. You can view the top 10 projected players according to ESPN and the top 10 actual fantasy leaders for the 2016 season below:

The first project at Metis involved analyzing the NYC Metropolitan Transportation Authority (MTA) turnstile data in order to provide recommendations to a womens’ tech organization as to which subway stations they should target in order to advertise for their annual gala.

Although the first week of the bootcamp was fast paced, I was pleasantly surprised to find that I knew more than I thought! This comfort was due to significant studying that I did independently in preparation for the bootcamp. In particular, I used the following resources that I highly recommend: