Over the weekend, I attended the D.C.-area Data Intelligence conference, which was organized by NumFocus. It was great to see some of the really interesting machine learning work that people are doing–certainly far too much to capture in a blog post. But check out the list of talks on the site linked above.

I recently completed a proof-of-concept project to see if it would be possible to predict traffic congestion in Washington, D.C. (where I live) based on political / governmental activity and overall volume of significant events.

I try to make a habit of bookmarking data sources that I come across. Even if I don’t use it for the project I’m working on, it’s handy to have a list at the ready.

For a recent project, I was scraping data from a few different websites and needed to solve for how to handle infinite scrolling. Here are two useful methods I found that worked for me. (They won’t work in every situation, but I think they are good to know.)

I recently completed a brief, but intense, team effort on an old Kaggle competition as an exercise in testing and sharpening our data science skills. We worked on the West Nile Prediction competition, which asks you to try and predict which mosquito traps in Chicago will contain a mosquito that tests positive for West Nile Virus.