Class-based approach to disambiguating Levin verbs

Lapata and Brew (Computational Linguistics, vol. 30, 2004, pp. 295–313) (hereafter LB04) obtain from untagged texts a statistical prior model that is able to generate class preferences for ambiguous Lewin (English Verb Classes and Alternations: A Preliminary Investigation, 1993, University of Chicago Press) verbs (hereafter Levin). They also show that their informative priors, incorporated into a Naive Bayes classifier deduced from hand-tagged data (HTD), can aid in verb class disambiguation. We re-analyse LB04's prior model and show that a single factor (the joint probability of class and frame) determines the predominant class for a particular verb in a particular frame. This means that the prior model cannot be sensitive to fine-grained lexical distinctions between different individual verbs falling in the same class. We replicate LB04's supervised disambiguation experiments on large-scale data, using deep parsers rather than the shallow parser of LB04. In addition, we introduce a method for training our classifier without using HTD. This relies on knowledge of Levin class memberships to move information from unambiguous to ambiguous instances of each class. We regard this system as unsupervised because it does not rely on human annotation of individual verb instances. Although our unsupervised verb class disambiguator does not match the performance of the ones that make use of HTD, it consistently outperforms the random baseline model. Our experiments also demonstrate that the informative priors derived from untagged texts help improve the performance of the classifier trained on untagged data.

[1]  P. Tichý.  Constructions , 1986, Philosophy of Science.

[2]  Mats Rooth,et al.  Valence Induction with a Head-Lexicalized PCFG , 1998, EMNLP.

[3]  George A. Miller,et al.  Using Corpus Statistics and WordNet Relations for Sense Identification , 1998, CL.

[4]  Dekang Lin,et al.  Automatic Retrieval and Clustering of Similar Words , 1998, ACL.

[5]  Chris Brew,et al.  Which Are the Best Features for Automatic Verb Classification , 2008, ACL.

[6]  David Yarowsky,et al.  Word-Sense Disambiguation Using Statistical Models of Roget’s Categories Trained on Large Corpora , 2010, COLING.

[7]  Anna Korhonen,et al.  Improving Subcategorization Acquisition Using Word Sense Disambiguation , 2003, ACL.

[8]  MerloPaola,et al.  Automatic verb classification based on statistical distributions of argument structure , 2001 .

[9]  Suzanne Stevenson,et al.  Automatic Verb Classification Based on Statistical Distributions of Argument Structure , 2001, CL.

[10]  Mirella Lapata,et al.  Constructing Semantic Space Models from Parsed Corpora , 2003, ACL.

[11]  Beth Levin,et al.  English Verb Classes and Alternations: A Preliminary Investigation , 1993 .

[12]  Sabine Schulte im Walde.  Clustering Verbs Semantically According to their Alternation Behaviour , 2000, COLING.

[13]  Douglas L. T. Rohde.  An Improved Method for Deriving Word Meaning from Lexical Co-Occurrence , 2004 .

[14]  Lei Shi,et al.  Putting Pieces Together: Combining FrameNet, VerbNet and WordNet for Robust Semantic Parsing , 2005, CICLing.

[15]  Hwee Tou Ng,et al.  An Empirical Evaluation of Knowledge Sources and Learning Algorithms for Word Sense Disambiguation , 2002, EMNLP.

[16]  James Henderson.  Inducing History Representations for Broad Coverage Statistical Parsing , 2003, HLT-NAACL.

[17]  Eugene Charniak,et al.  A Maximum-Entropy-Inspired Parser , 2000, ANLP.

[18]  Ted Dunning,et al.  Accurate Methods for the Statistics of Surprise and Coincidence , 1993, CL.

[19]  Frank Keller,et al.  Finding Syntactic Structure in Unparsed Corpora , 2001 .

[20]  E. Rosch,et al.  Family resemblances: Studies in the internal structure of categories , 1975, Cognitive Psychology.

[21]  Mirella Lapata,et al.  Using Semantic Roles to Improve Question Answering , 2007, EMNLP.

[22]  van Gerardus Noord,et al.  Special issue: finite state methods in natural language processing , 2003 .

[23]  Hinrich Schütze,et al.  Automatic Word Sense Discrimination , 1998, Comput. Linguistics.

[24]  Georgia M. Green,et al.  Semantics and Syntactic Regularity , 1974 .

[25]  Mirella Lapata,et al.  Verb Class Disambiguation Using Informative Priors , 2004, CL.

[26]  Suzanne Stevenson,et al.  A General Feature Space for Automatic Verb Classification , 2003, EACL.

[27]  Martha Palmer,et al.  Investigating Regular Sense Extensions Based on Intersective Levin Classes , 1998, COLING-ACL.

[28]  Yuval Krymolowski,et al.  Clustering Polysemic Subcategorization Frame Distributions Semantically , 2003, ACL.

[29]  Julie Weeds,et al.  Finding Predominant Word Senses in Untagged Text , 2004, ACL.

[30]  Frank Keller,et al.  Finding Syntactic Structure in Unparsed Corpora The Gsearch Corpus Query System , 2001, Comput. Humanit..

[31]  Maria Lapata,et al.  Acquiring Lexical Generalizations from Corpora: A Case Study for Diathesis Alternations , 1999, ACL.

[32]  Walter Daelemans,et al.  Parameter optimization for machine-learning of word sense disambiguation , 2002, Natural Language Engineering.

[33]  Martha Palmer,et al.  Class-Based Construction of a Verb Lexicon , 2000, AAAI/IAAI.

[34]  Martha Palmer,et al.  Consistent Criteria for Sense Distinctions , 2000, Comput. Humanit..

[35]  Suzanne Stevenson,et al.  Unsupervised Semantic Role Labellin , 2004, EMNLP.

[36]  John A. Carroll,et al.  Applied morphological processing of English , 2001, Natural Language Engineering.

[37]  Daniel Jurafsky,et al.  Automatic Labeling of Semantic Roles , 2002, CL.

[38]  David R. Dowty.  Thematic proto-roles and argument selection , 1991 .

[39]  Walter Daelemans,et al.  Classifier Optimization and Combination in the English All Words Task , 2001, *SEMEVAL.