In my last blog post I spoke about my worry that the fifth week of the bootcamp would be uninteresting due to my previous exposure to Python. Sadly, that mostly turned out to be the case. The week consisted mostly of going over the basics of the language with an introduction to web scraping, regular expressions, and the IPython notebook thrown in. It took until Friday for my curiosity to wake up as we received the first lecture in a two part series on Natural Language Processing (NLP). The first class was an introduction, setting the theoretical foundation needed for us to arroach the subject computationally. Our instructor studied linguistics, and his enthusiasm for the material was infectious. (At least to me it was, but I suppose I’m biased because I’m also interested in linguistics).
My previous exposure to NLP was a bit overwhelming. It was too brief for me to do much other than run code blindly, and I never did revisit it due to prioritizing other fundamental aspects of Data Science. This unexpected rendevous has rekindled that flame of fascination, and I plan to explore the topic more. In fact, I already have a couple of ideas bouncing around.
Another important section of Week 6 was the presentation of several Shiny projects. It’s always interesting to see ideas that the other members of the cohort produced, and a couple of them really blew me away. My effort centered around The Death Penalty Information Center’s database of those executed in the United States since 1977. In essence, it is sort of a wrapper with the added perks of the data being accompanied by explorable, real time visualizations. I’ll get around to talking about it here soon.
Week 6 brings more Python as we dive more into the language’s Data Science stack with introductions to numpy, pandas, and scipy, for example.