Paper: Sentiment Translation through Multi-Edge Graphs

ACL ID C10-2127
Title Sentiment Translation through Multi-Edge Graphs
Venue International Conference on Computational Linguistics
Session Poster Session
Year 2010

Sentiment analysis systems can benefit fromthetranslationofsentimentinforma- tion. We present a novel, graph-based ap- proach using SimRank, a well-established graph-theoretic algorithm, to transfer sen- timent information from a source lan- guage to a target language. We evaluate this method in comparison with semantic orientation using pointwise mutual infor- mation(SO-PMI), anestablished unsuper- vised method for learning the sentiment of phrases.