FeasPar: a feature structure parser learning to parse spoken language

Authors:
Finn Dag Buø;Alex Waibel
Affiliations:
University of Karlsruhe, Germany, and Carnegie Mellon University;University of Karlsruhe, Germany, and Carnegie Mellon University
Venue:
COLING '96 Proceedings of the 16th conference on Computational linguistics - Volume 1
Year:
1996

Citing 5
Cited 0

A statistical approach to machine translation

Computational Linguistics
TINA: a natural language system for spoken language applications

Computational Linguistics
Parsec: a connectionist learning architecture for parsing spoken language

Parsec: a connectionist learning architecture for parsing spoken language
Learning fault-tolerant speech parsing with SCREEN

AAAI '94 Proceedings of the twelfth national conference on Artificial intelligence (vol. 1)
Gemini: a natural language system for spoken-language understanding

HLT '93 Proceedings of the workshop on Human Language Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

We describe and experimentally evaluate a system, FeasPar, that learns parsing spontaneous speech. To train and run FeasPar (Feature Structure Parser), only limited handmodeled knowledge is required.The FeasPar architecture consists of neural networks and a search. The networks split the incoming sentence into chunks, which are labeled with feature values and chunk relations. Then, the search finds the most probable and consistent feature structure.FeasPar is trained, tested and evaluated with the Spontaneous Scheduling Task, and compared with a handmodeled LR-parser. The handmodeling effort for FeasPar is 2 weeks. The handmodeling effort for the LR-parser was 4 months. FeasPar performed better than the LR-parser in all six comparisons that are made.