Identification of Subject Shareness for Korean-English Machine Translation

Authors:
Kye-Sung Kim;Seong-Bae Park;Hyun-Je Song;Se-Young Park;Sang-Jo Lee
Affiliations:
Department of Computer Engineering, Kyungpook National University, Daegu, Korea 702-701;Department of Computer Engineering, Kyungpook National University, Daegu, Korea 702-701;Department of Computer Engineering, Kyungpook National University, Daegu, Korea 702-701;Department of Computer Engineering, Kyungpook National University, Daegu, Korea 702-701;Department of Computer Engineering, Kyungpook National University, Daegu, Korea 702-701
Venue:
PRICAI '08 Proceedings of the 10th Pacific Rim International Conference on Artificial Intelligence: Trends in Artificial Intelligence
Year:
2008

Citing 8
Cited 1

Centering: a framework for modeling the local coherence of discourse

Computational Linguistics
Making large-scale support vector machine learning practical

Advances in kernel methods
Pronominal Anaphora Generation in an English-Spanish MT Approach

CICLing '02 Proceedings of the Third International Conference on Computational Linguistics and Intelligent Text Processing
Interlingua-based broad-coverage Korean-to-English translation in CCLINC

HLT '01 Proceedings of the first international conference on Human language technology research
Phrase-pattern-based Korean to English machine translation using two level translation pattern selection

ACL '00 Proceedings of the 38th Annual Meeting on Association for Computational Linguistics
Japanese zero pronoun resolution based on ranking rules and machine learning

EMNLP '03 Proceedings of the 2003 conference on Empirical methods in natural language processing
Two-Phase S-Clause Segmentation

IEICE - Transactions on Information and Systems
Clause restructuring for statistical machine translation

ACL '05 Proceedings of the 43rd Annual Meeting on Association for Computational Linguistics

Identification of non-referential zero pronouns for Korean-English machine translation

PRICAI'10 Proceedings of the 11th Pacific Rim international conference on Trends in artificial intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

One of the most critical issues in translating Korean into other languages is the common use of empty arguments. Since even mandatory elements in Korean are often dropped unlike English, the missing elements should be resolved during translation to obtain grammatical sentences. In this paper, we focus on missing subjects in intra-sentential level, which can be regarded as the identification of subject sharing between clauses. In order to reflect syntactic information in resolving missing subjects, we use a parse tree kernel, a specialized convolution kernel. In experimental evaluation, syntactic information turns out to be positively related to the identification of subject shareness. Our method achieves an accuracy of 81.39% and outperforms the baseline system assuming that two adjacent clauses share a subject.