Summary of the paper
| Title | A Semi-supervised Type-based Classification of Adjectives: Distinguishing Properties and Relations |
|---|---|
| Authors | Matthias Hartung and Anette Frank |
| Abstract | We present a semi-supervised machine-learning approach for the classificationof adjectives into property- vs. relation-denoting adjectives, a distinctionthat is highly relevant for ontology learning. The feasibility of thisclassification taskis evaluated in a human annotation experiment. We observe that token-levelannotation of these classes is expensive and difficult. Yet, a careful corpusanalysis reveals that adjective classes tend to be stable, with few occurrencesof class shifts observed at the token level. As a consequence, we opt for atype-based semi-supervised classification approach. The class labels obtainedfrom manual annotation are projected to large amounts of unannotated tokensamples. Training on heuristically labeled data yields high classificationperformance on our own data and on a data set compiled from WordNet. Ourresults suggest that it is feasible to automatically distinguish adjectivesdenoting properties and relations, using small amounts of annotated data. |
| Language | Ontologies |
| Topics | Corpus (creation, annotation, etc.), Statistical and machine learning methods, Ontologies |
Full paper  |
A Semi-supervised Type-based Classification of Adjectives: Distinguishing Properties and Relations |
| Bibtex | @InProceedings{HARTUNG10.685, author = {Matthias Hartung and Anette Frank}, title = {A Semi-supervised Type-based Classification of Adjectives: Distinguishing Properties and Relations}, 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} } |