This was my senior independent study project at the Hill School - analyzing hateful speech and negative sentiments of Tweets made during the 2016 and 2020 presidential elections.
I used the Decision Tree Classifier algorithm by SciKit Learn to detect and classify hateful speech in almost 2 million Twitter posts relating to the 2016 and 2020 presidential elections. The DTC was trained using a sample dataset of 32,000 tweets classified as either “neutral,” labeled “0,” or “hateful,” labeled “1.”