Analyzing sentiment while accounting for negation scope and strength

Recent developments in automated sentiment analysis show a tendency of accounting for various aspects other than word frequencies. One of these aspects is negation. We compare several approaches to accounting for negation in sentiment analysis, differing in their methods of determining the scope of influence of a negation keyword. On a set of English movie review sentences, the best approach turns out to be to consider the first two words, following a negation keyword, to be negated by that keyword. Additionally, we propose to optimize the sentiment modification in case of negation to a value of ‐1.27 rather than ‐1.