Skip to main navigation menu Skip to main content Skip to site footer

Using Pivot-Based Paraphrasing and Sentiment Profiles to Improve a Subjectivity Lexicon for Essay Data

Abstract

We demonstrate a method of improving a seed sentiment lexicon developed on essay data by using a pivot-based paraphrasing system for lexical expansion coupled with sentiment profile enrichment using crowdsourcing. Profile enrichment alone yields up to 15% improvement in the accuracy of the seed lexicon on 3-way sentence-level sentiment polarity classification of essay data. Using lexical expansion in addition to sentiment profiles provides a further 7% improvement in performance. Additional experiments show that the proposed method is also effective with other subjectivity lexicons and in a different domain of application (product reviews). 

PDF (Presented at ACL 2013)

Author Biography

Beata Beigman Klebanov

  Beata Beigman Klebanov is a research scientist in the Research and Development division at Educational Testing Service in Princeton, NJ.  She received her Ph.D. in computer science (computational linguistics) in 2008 and her B.S. degree (Magna Cum Laude) in computer science in 2000—both from The Hebrew University of Jerusalem, Israel.  She received her M.S. degree (with distinction) in cognitive science from the University of Edinburgh in 2001.

 Beigman Klebanov works in the Natural Language Processing and Speech Group in the Assessment Innovations Center specializing in Natural Language Processing.  Before joining ETS, she was a post-doctoral fellow at the Northwestern Institute for Complex Systems and Kellogg School of Management where she researched computational approaches to political rhetoric.  Her interests include discourse modeling, analyzing argumentative and figurative language, and automated semantic and pragmatic analysis of text.  At ETS, her focus is on automatically scoring content in student writing.  She researches methods to analyze cohesion in student essays, metaphor and sentiment, among others.