This was a project done for EECS 280, one of UMich's intro programming classes.

We were tasked with training a machine learning model on posts from an online classroom forum labeled with their topics. We then needed to use the model to predict the topic of a set of posts it had never seen before. We used a strategy called the bag of words, which looks for the number of occurrences of specific words by topic, and uses it as a measurement of how well a post fits a topic.