Paper: Learning with Compositional Semantics as Structural Inference for Subsentential Sentiment Analysis

ACL ID D08-1083
Title Learning with Compositional Semantics as Structural Inference for Subsentential Sentiment Analysis
Venue Conference on Empirical Methods in Natural Language Processing
Session Main Conference
Year 2008
Authors

Determining the polarity of a sentiment- bearing expression requires more than a sim- ple bag-of-words approach. In particular, words or constituents within the expression can interact with each other to yield a particu- laroverallpolarity. Inthispaper, weviewsuch subsentential interactions in light of composi- tionalsemantics,and presentanovel learning- based approach that incorporatesstructuralin- ference motivated by compositional seman- tics into the learning procedure. Our exper- iments show that (1) simple heuristics based on compositional semantics can perform bet- terthanlearning-basedmethodsthatdonotin- corporate compositional semantics (accuracy of 89.7% vs. 89.1%), but (2) a method that integrates compositional semantics into learn- ing performs better than all other alterna- ti...