Improving unsupervised vector-space thematic fit evaluation via role-filler prototype clustering

Most recent unsupervised methods in vector space semantics for assessing thematic fit (e.g. Erk, 2007; Baroni and Lenci, 2010; Sayeed and Demberg, 2014) create prototypical rolefillers without performing word sense disambiguation. This leads to a kind of sparsity problem: candidate role-fillers for different senses of the verb end up being measured by the same “yardstick”, the single prototypical role-filler. In this work, we use three different feature spaces to construct robust unsupervised models of distributional semantics. We show that correlation with human judgements on thematic fit estimates can be improved consistently by clustering typical role-fillers and then calculating similarities of candidate rolefillers with these cluster centroids. The suggested methods can be used in any vector space model that constructs a prototype vector from a non-trivial set of typical vectors.

[1]  Frank Keller,et al.  A Probabilistic Model of Semantic Plausibility in Sentence Processing , 2009, Cogn. Sci..

[2]  Stefan Thater,et al.  Word Meaning in Context: A Simple and Effective Vector Model , 2011, IJCNLP.

[3]  Yuji Matsumoto.  MaltParser: A language-independent system for data-driven dependency parsing , 2005 .

[4]  Raymond J. Mooney,et al.  Multi-Prototype Vector-Space Models of Word Meaning , 2010, NAACL.

[5]  M. Tanenhaus,et al.  Modeling the Influence of Thematic Fit (and Other Constraints) in On-line Sentence Comprehension , 1998 .

[6]  Mirella Lapata,et al.  A Comparison of Vector-based Representations for Semantic Composition , 2012, EMNLP.

[7]  Ulrike Padó,et al.  The integration of syntax and semantic plausibility in a wide-coverage model of human sentence processing , 2007 .

[8]  Katrin Erk,et al.  Vector Space Models of Word Meaning and Phrase Meaning: A Survey , 2012, Lang. Linguistics Compass.

[9]  Marco Baroni,et al.  BagPack: A General Framework to Represent Semantic Relations , 2009, ArXiv.

[10]  T. Caliński,et al.  A dendrite method for cluster analysis , 1974 .

[11]  Ewan Klein,et al.  Natural Language Processing with Python , 2009 .

[12]  K. Rayner,et al.  Contextual effects on word perception and eye movements during reading , 1981 .

[13]  Andrew Y. Ng,et al.  Improving Word Representations via Global Context and Multiple Word Prototypes , 2012, ACL.

[14]  Katrin Erk,et al.  A Simple, Similarity-based Model for Selectional Preferences , 2007, ACL.

[15]  Dimitri Kartsaklis,et al.  Resolving Lexical Ambiguity in Tensor Regression Models of Meaning , 2014, ACL.

[16]  Christopher D. Manning,et al.  Generating Typed Dependency Parses from Phrase Structure Parses , 2006, LREC.

[17]  K. McRae,et al.  Integrating Verbs, Situation Schemas, and Thematic Role Concepts , 2001 .

[18]  Silvia Bernardini,et al.  Introducing and evaluating ukWaC , a very large web-derived corpus of English , 2008 .

[19]  Mats Rooth,et al.  Inducing a Semantically Annotated Lexicon via EM-Based Clustering , 1999, ACL.

[20]  Walter Daelemans,et al.  A Robust and Extensible Exemplar-Based Model of Thematic Fit , 2009, EACL.

[21]  Jason Weston,et al.  Natural Language Processing (Almost) from Scratch , 2011, J. Mach. Learn. Res..

[22]  Alessandro Lenci,et al.  Distributional Memory: A General Framework for Corpus-Based Semantics , 2010, CL.

[23]  Katrin Erk,et al.  A Flexible, Corpus-Driven Model of Regular and Inverse Selectional Preferences , 2010, CL.

[24]  Clayton Greenberg.  Disambiguating prepositional phrase attachment sites with sense information captured in contextualized distributional data , 2014, ACL.

[25]  Bernd Bohnet,et al.  Very high accuracy and fast dependency parsing is not a contradiction , 2010, COLING 2010.

[26]  Andrew McCallum,et al.  Efficient Non-parametric Estimation of Multiple Embeddings per Word in Vector Space , 2014, EMNLP.

[27]  P. Resnik.  Selectional constraints: an information-theoretic model and its computational realization , 1996, Cognition.