Using the labeled data, we further design and evaluate the performance of classifiers to automatically detect and predict incidents of cyberbullying and cyberaggression.