Richard Collins

• Used Semeval 2017 datasets to build three sentiment classifiers capable of classifying tweets into positive, neutral, and negative classes. • Leveraged tweet features and applied random forest models with tuned parameters to fit training data, and predict sentiment labels.