Summary of the paper
| Title | Training Parsers on Partial Trees: A Cross-language Comparison |
|---|---|
| Authors | Kathrin Spreyer, Lilja Øvrelid and Jonas Kuhn |
| Abstract | We present a study that compares data-driven dependency parsers obtained bymeans of annotation projection between language pairs of varying structural similarity. We show how the partial dependency treesprojected from English to Dutch, Italian and German can be exploited to train parsers for the target languages. We evaluate the parsersagainst manual gold standard annotations and find that the projected parsers substantially outperform our heuristic baselines by 9―25%UAS, which corresponds to a 21―43% reduction in error rate. A comparative error analysis focuses on how the projected target languageparsers handle subjects, which is especially interesting for Italian as an instance of a pro-drop language. For Dutch, we further presentexperiments with German as an alternative source language. In both source languages, we contrast standard baseline parsers with parsersthat are enhanced with the predictions from large-scale LFG grammars through a technique of parser stacking, and show that improvements ofthe source language parser can directly lead to similar improvements of the projected target language parser. |
| Language | Grammar and Syntax |
| Topics | Parsing, Multilinguality, Grammar and Syntax |
Full paper  |
Training Parsers on Partial Trees: A Cross-language Comparison |
| Bibtex | @InProceedings{SPREYER10.722, author = {Kathrin Spreyer, Lilja Øvrelid and Jonas Kuhn}, title = {Training Parsers on Partial Trees: A Cross-language Comparison}, 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} } |