Summary of the paper
| Title | An Evaluation of Predicate Argument Clustering using Pseudo-Disambiguation |
|---|---|
| Authors | Christian Scheible |
| Abstract | Schulte im Walde et al. (2008) presented a novel approach to semantic verbclassication. The predicate argument model (PAC) presented in their papermodels selectional preferences by using soft clustering that incorporates theExpectation Maximization (EM) algorithm and the MDL principle. In this paper, Iwill show how the model handles the task of differentiating between plausibleand implau- sible combinations of verbs, subcategorization frames and argumentsby applying the pseudo-disambiguation evaluation method. The predicate argumentclustering model will be evaluated in comparison with the latent semanticclustering model by Rooth et al. (1999). In particular, the influences of themodel parameters, data frequency, and the individual components of thepredicate argument model are examined. The results of these experiments showthat (i) the selectional preference model overgeneralizes over arguments forthe purpose of a pseudo-disambiguation task and that (ii) pseudo-disambiguationshould not be used as a universal indicator for the quality of a model. |
| Language | Evaluation methodologies |
| Topics | Language modelling, Statistical and machine learning methods, Evaluation methodologies |
Full paper  |
An Evaluation of Predicate Argument Clustering using Pseudo-Disambiguation |
| Bibtex | @InProceedings{SCHEIBLE10.206, author = {Christian Scheible}, title = {An Evaluation of Predicate Argument Clustering using Pseudo-Disambiguation}, booktitle = {Proceedings of the Seventh conference on International Language Resources and Evaluation (LREC'10)}, year = {2010}, month = {may}, date = {19-21}, address = {Valletta, Malta}, editor = {Nicoletta Calzolari (Conference Chair), Khalid Choukri, Bente Maegaard, Joseph Mariani, Jan Odjik, Stelios Piperidis, Mike Rosner, Daniel Tapias}, publisher = {European Language Resources Association (ELRA)}, isbn = {2-9517408-6-7}, language = {english} } |