Acquiring Verb Subcategorization Frames in Bengali from Corpora

Authors:
Dipankar Das;Asif Ekbal;Sivaji Bandyopadhyay
Affiliations:
Department of Computer Science and Engineering, Jadavpur University, Kolkata, India;Department of Computer Science and Engineering, Jadavpur University, Kolkata, India;Department of Computer Science and Engineering, Jadavpur University, Kolkata, India
Venue:
ICCPOL '09 Proceedings of the 22nd International Conference on Computer Processing of Oriental Languages. Language Technology for the Knowledge-based Economy
Year:
2009

Citing 5
Cited 0

Large lexicons for natural language processing: utilising the grammar coding system of LDOCE

Computational Linguistics - Special issue of the lexicon
Automatic acquisition of a large subcategorization dictionary from corpora

ACL '93 Proceedings of the 31st annual meeting on Association for Computational Linguistics
Comlex Syntax: building a computational lexicon

COLING '94 Proceedings of the 15th conference on Computational linguistics - Volume 1
Automatic extraction of subcategorization frames for Czech

COLING '00 Proceedings of the 18th conference on Computational linguistics - Volume 2
A Data mining approach for resolving cases of Multiple Parsing in Machine Aided Translation of Indian Languages

ITNG '07 Proceedings of the International Conference on Information Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

Subcategorization frames acquisition of a phrase can be described as a mechanism to extract different types of relevant arguments that are associated with that phrase in a sentence. This paper presents the acquisition of different subcategory frames for a specific Bengali verb that has been identified from POS tagged and chunked data prepared from raw Bengali news corpus. Syntax plays the main role in the acquisition process and not the semantics like thematic roles. The output frames of the verb have been compared with the frames of its English verb that has been identified using bilingual lexicon. The frames for the English verb have been extracted using Verbnet. This system has demonstrated precision and recall values of 85.21% and 83.94% respectively on a test set of 1500 sentences.