Modeling Perspective Using Adaptor Grammars

Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pages 284–292, MIT, Massachusetts, USA, 9-11 October 2010. c⃝2010 Association for Computational Linguistics Modeling Perspective using Adaptor Grammars Eric A. Hardisty Department of Computer Science and UMIACS University of Maryland College Park, MD hardisty@cs.umd.edu Jordan Boyd-Graber UMD iSchool and UMIACS University of Maryland College Park, MD jbg@umiacs.umd.edu Philip Resnik Department of Linguistics and UMIACS University of Maryland College Park, MD resnik@umd.edu Abstract Strong indications of perspective can often come from collocations of arbitrary length; for example, someone writing get the government out of my X is typically expressing a conserva- tive rather than progressive viewpoint. How- ever, going beyond unigram or bigram features in perspective classification gives rise to prob- lems of data sparsity. We address this prob- lem using nonparametric Bayesian modeling, specifically adaptor grammars (Johnson et al., 2006). We demonstrate that an adaptive na¨ıve Bayes model captures multiword lexical usages associated with perspective, and establishes a new state-of-the-art for perspective classifica- tion results using the Bitter Lemons corpus, a collection of essays about mid-east issues from Israeli and Palestinian points of view. 1 Introduction Most work on the computational analysis of senti- ment and perspective relies on lexical features. This makes sense, since an author’s choice of words is often used to express overt opinions (e.g. describing healthcare reform as idiotic or wonderful) or to frame a discussion in order to convey a perspective more implicitly (e.g. using the term death tax instead of estate tax). Moreover, it is easy and efficient to rep- resent texts as collections of the words they contain, in order to apply a well known arsenal of supervised techniques (Laver et al., 2003; Mullen and Malouf, 2006; Yu et al., 2008). At the same time, standard lexical features have their limitations for this kind of analysis. Such fea- tures are usually created by selecting some small n-gram size in advance. Indeed, it is not uncommon to see the feature space for sentiment analysis limited to unigrams. However, important indicators of per- spective can also be longer (get the government out of my). Trying to capture these using standard ma- chine learning approaches creates a problem, since allowing n-grams as features for larger n gives rise to problems of data sparsity. In this paper, we employ nonparametric Bayesian models (Orbanz and Teh, 2010) in order to address this limitation. In contrast to parametric models, for which a fixed number of parameters are specified in advance, nonparametric models can “grow” to the size best suited to the observed data. In text analysis, models of this type have been employed primarily for unsupervised discovery of latent structure — for example, in topic modeling, when the true number of topics is not known (Teh et al., 2006); in grammatical inference, when the appropriate number of nontermi- nal symbols is not known (Liang et al., 2007); and in coreference resolution, when the number of enti- ties in a given document is not specified in advance (Haghighi …

Content truncated to 500 words