Sentiment Analysis of Short Informal Texts

• DOI: 10.1613/jair.4272
• Publication Date: 2018-07-18T10:49:56Z
• Authors: S. Kiritchenko; X. Zhu; S. M. Mohammad
• Keywords: SemEval; Sentiment Analysis; Phrase

We describe a state-of-the-art sentiment analysis system that detects (a) the of short informal textual messages such as tweets and SMS (message-level task) (b) word or phrase within message (term-level task). The is based on supervised statistical text classification approach leveraging variety surface-form, semantic, features. features are primarily derived from novel high-coverage tweet-specific lexicons. These lexicons automatically generated with sentiment-word hashtags emoticons. To adequately capture words in negated contexts, separate lexicon for words. ranked first SemEval-2013 shared task `Sentiment Analysis Twitter' (Task 2), obtaining an F-score 69.02 message-level 88.93 term-level task. Post-competition improvements boost performance to 70.45 89.50 also obtains two additional datasets: test set corpus movie review excerpts. ablation experiments demonstrate use results gains up 6.5 absolute percentage points.