Abstract | ||
---|---|---|
We describe a novel method for coping with ungrammatical input based on the use of chart-like data structures, which permit anytime processing. Priority is given to deep syntactic analysis. Should this fail, the best partial analyses are selected, according to a shortest-paths algorithm, and assembled in a robust processing phase. The method has been applied in a speech translation project with large HPSG grammars. |
Year | DOI | Venue |
---|---|---|
1999 | 10.3115/1034678.1034741 | ACL |
Keywords | Field | DocType |
chart-like data structure,large hpsg grammar,novel method,speech translation project,robust speech parsing,ungrammatical input,robust processing phase,shortest-paths algorithm,deep syntactic analysis,partial analysis,data structure,shortest path algorithm | Head-driven phrase structure grammar,Rule-based machine translation,Data structure,Computer science,Speech recognition,Natural language processing,Artificial intelligence,Parsing,Speech translation,Machine learning | Conference |
Volume | ISBN | Citations |
P99-1 | 1-55860-609-3 | 14 |
PageRank | References | Authors |
2.31 | 7 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
W. Kasper | 1 | 14 | 2.31 |
Bernd Kiefer | 2 | 194 | 25.67 |
H. U. Krieger | 3 | 131 | 14.70 |
C. J. Rupp | 4 | 67 | 11.66 |
K. L. Worm | 5 | 14 | 2.31 |