Here is my Jupyter notebook and slide show PDF for my work on the OkCupid data set. I attempted to build optimum models to predict those in the data set that use longer than average words. I considered that this could be fun because those using those types of words may be well-read but also could be a bit pretentious. I chose a varied feature set, algorithm types, and algorithm parameters to attempt to predict. I had fun with this one and wanted to share. I could keep working at this one with new feature columns.