TwitterHawk is an open-source natural language processing system for Twitter sentiment analysis. This system was developed for the SemEval-2015 Task 10: Sentiment Analysis in Twitter.

The system placed 1st in topic-based sentiment classification (Subtask C), and ranked 6th out of 40 in identifying the sentiment of sarcastic tweets (for Subtask B). The system also placed 5th/11 for Subtask A (sentiment of a tweet's sub-phrase), 10th/40 for Subtask B (sentiment of a full tweet), and 3rd/6 for Subtask D (summarization of Subtask C).

TwitterHawk implements two classifiers, one for message-level sentiment, and one of phrase-level sentiment detection.

  • Free software: Apache v2.0 license

View on GitHub


W. Boag, P. Potash, A. Rumshisky. TwitterHawk: A Feature Bucket Approach to Sentiment Analysis. The 9th International Workshop on Semantic Evaluation (SemEval-2015). NAACL HLT 2015, May 31-Jun 5. Dever, Colorado.