If Elon Musk tweets in the forest and no one is there to...? Whatever. In one of my previous posts I tried to distinguish between tweets from @BoredElonMusk and @elonmusk using a Naive Bayes Classifier. Now, I attempt the same exercise but using a Random Forest Classier. Let's see which one can make a better prediction...
If you're Canadian and live in Ontario, it was pretty much a national duty to cheer for the Toronto Raptors during the finals against Golden State. To further jump on this bandwagon, I decided to do a network analysis of the passes made by Toronto Raptors players in the finals. I use an unoffical NBA API for data and the networkx package.
In this post, I use Multinomial Naive Bayes to classify tweets as the "real" or "fake" Elon Musk. Words are filtered from each tweet, transformed into term frequency-inverse document frequency (TF-IDF) space and used as features in the model.
In this post, I replicate a Python Data Science Handbook example that predicts bike traffic counts. The example in the book uses Seattle data, whereas I use bike counts from the Galloping Goose Trail in Victoria, BC.
In this post, I discuss how to install Anaconda on Mac and use Jupyter Notebooks. I mainly wanted to document them for myself in case I have to do this again.
© Monica Mow 2019
Powered by Pelican