The final project will be based on attempts to solve and submissions to the Spooky Author Identification contest.
The contest is defined at the page linked above. The Final project will be to show at least three different submissions and their scores. You should show your work in jupyter notebooks as usual. Your grade will be based on the quality of your description of what you did and your analysis of the results.
You may submit differnt notebooks for each of your attempts or you can combine them into a single notebook.
You may find it useful to look at the Natural Language Tollkit (NLTK) as many of the existing kernels use this.
You are also welcome to look at any existing kernels published for the contest and I encourage you to submit your own. Be careful, as this can be a rabbit hole in which you waste a lot of time trying to understand what someone is doing without actually moving forward with your work.
Also I encourage you to ask questions on the forum for this contest.
Get started early and ask questions in class as you go. This is a really interesting project and should be a lot of fun.