Multi-word expression recognition integrated with two-level finite state transducer

Authors:
Keunyong Lee;Ki-Soen Park;Yong-Seok Lee
Affiliations:
Division of Electronics and Information Engineering, Chonbuk National University, Jeonju, South Korea;Division of Electronics and Information Engineering, Chonbuk National University, Jeonju, South Korea;Division of Electronics and Information Engineering, Chonbuk National University, Jeonju, South Korea
Venue:
HCI'07 Proceedings of the 12th international conference on Human-computer interaction: intelligent multimodal interaction environments
Year:
2007

Citing 8
Cited 0

Theory of finite automata with an introduction to formal languages

Theory of finite automata with an introduction to formal languages
Compiler design in C

Compiler design in C
Multiword Expressions: A Pain in the Neck for NLP

CICLing '02 Proceedings of the Third International Conference on Computational Linguistics and Intelligent Text Processing
The replace operator

ACL '95 Proceedings of the 33rd annual meeting on Association for Computational Linguistics
Directed replacement

ACL '96 Proceedings of the 34th annual meeting on Association for Computational Linguistics
Constructing lexical transducers

COLING '94 Proceedings of the 15th conference on Computational linguistics - Volume 1
Engineering A Compiler

Engineering A Compiler
Integrating morphology with multi-word expression processing in Turkish

MWE '04 Proceedings of the Workshop on Multiword Expressions: Integrating Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes another two-level finite state transducer to recognize the multi-word expression (MWE) in two-level morphological parsing environment. In our proposed the Finite State Transducer with Bridge State (FSTBS), we defined Bridge State (concerned with connection of multi-word), Bridge Character (used in connection of multi-word expression) and two-level rule to extend existing FST. FSTBS could recognize both Fixed Type MWE and Flexible Type MWE which are expressible as regular expression, because FSTBS recognizes MWE in morphological parsing.