Two languages are better than one (for syntactic parsing)

Burkett, David; Klein, Dan

doi:10.3115/1613715.1613828

Cited by 39 publications

(39 citation statements)

References 16 publications

(13 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For efficiency, we used 2K samples from the training corpus and set the mini-batch size in feature-enhanced EM to 200 similar to "rapid training" in (Burkett and Klein, 2008). We also set µ b = 50 during EM training to manage the training time.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Monolingual Phrase Alignment on Parse Forests

Arase

Tsujii

2017

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

We propose an efficient method to conduct phrase alignment on parse forests for paraphrase detection. Unlike previous studies, our method identifies syntactic paraphrases under linguistically motivated grammar. In addition, it allows phrases to non-compositionally align to handle paraphrases with non-homographic phrase correspondences. A dataset that provides gold parse trees and their phrase alignments is created. The experimental results confirm that the proposed method conducts highly accurate phrase alignment compared to human performance.

show abstract

Section: Resultsmentioning

confidence: 99%

“…Also, mini-batch training (Liang and Klein, 2009) is applied. Such an approximation for efficiency is common in parallel parsing (Burkett and Klein, 2008;Burkett et al, 2010).…”

Section: Feature-enhanced Em Algorithmmentioning

confidence: 99%

Monolingual Phrase Alignment on Parse Forests

Arase

Tsujii

2017

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

show abstract

“…Burkett and Klein (2008) firstly proposed to jointly train on translated treebanks of two languages, which leads to improvements in both parsing accuracy and MT. Huang, Jiang, and Liu (2009) improved the work of (Burkett and Klein 2008), in which reordering information acquired by a source parser is used for bilingual parsing. Zhao, Song, Kit, and Zhou (2009) proposed to use a bilingual lexicon for bilingual parsing.…”

Section: Projection For Improving Parsing Accuracymentioning

confidence: 99%

Constrained Partial Parsing Based Dependency Tree Projection for Tree-to-Tree Machine Translation

Chu

Shen

Cromierès

et al. 2017

Journal of Natural Language Processing

View full text Add to dashboard Cite

Ideally, tree-to-tree machine translation (MT) that utilizes syntactic parse trees on both source and target sides could preserve non-local structure, and thus generate fluent and accurate translations. In practice, however, firstly, high quality parsers for both source and target languages are difficult to obtain; secondly, even if we have high quality parsers on both sides, they still can be non-isomorphic because of the annotation criterion difference between the two languages. The lack of isomorphism between the parse trees makes it difficult to extract translation rules. This extremely limits the performance of tree-to-tree MT. In this article, we present an approach that projects dependency parse trees from the language side that has a high quality parser, to the side that has a low quality parser, to improve the isomorphism of the parse trees. We first project a part of the dependencies with high confidence to make a partial parse tree, and then complement the remaining dependencies with partial parsing constrained by the already projected dependencies. Experiments conducted on the Japanese-Chinese and English-Chinese language pairs show that our proposed method significantly improves the performance on both the two language pairs.

show abstract

“…In many societies of Europe where the monolingual is used more than multilingual, these tendencies are also strong. In education, culture, social life, and employment, the knowledge of one language is unavoidable (Aronin & Singleton, 2008;Burkett & Klein, 2008;Cook, 1993;Graddol, 1997).…”

Section: Review Of Literaturementioning

confidence: 99%

The Study of the Effect of Educational Level of Students and English Language Teachers’ Awareness on True and False Cognates in Iran

Gholami¹,

Alavinia²,

Izadpanah³

2015

SUET

View full text Add to dashboard Cite

The ability to speak two languages in the non-English-speaking countries is a remarkable achievement. There is a good reason to believe that bilingualism is the norm for the majority of people in the world, because 70% of the earth's population are supposed to be bilingual or multilingual. Our purpose of the study was to determine the participants' awareness of using true and false cognate words of L1 (the first language) & L2 (the second language) in learning English. Three hundred and eighty-five selected people from 3,789 statistical population participated in 2014-2015 at three levels of A.D., B.A., and M.A. of Islamic Azad University as well as teachers of English in English educational institutions in Zanjan (Iran). Each educational level was considered as one category with the total sample calculated by using Kokran Formula, and the amount of each category was determined by using appropriate proportion and randomized categorical sampling method. The participants' age was between from 18 to 52, with a mean age of 29 years. The materials were 45 words of true and false cognate's words from 500 words by doing CVR (content validity ratio) and CVI (content validity index) (Lawshe's table with the index of 88% and 82% respectively) for being reliable and valid. ANOVA (Analysis of variance) was used for studying the effect of educational level on the rate of awareness. In other words, the awareness level has been compared in different educational levels. The results of Dankan test show that: There is no significant difference between the awareness of A.D., B.A., and M.A. levels, but there is a significant difference between the level of awareness of teachers group and the other groups (in true cognate words) and for false cognate words, there is no significant difference between the level of awareness of the four groups in this regard.

show abstract

Two languages are better than one (for syntactic parsing)

Cited by 39 publications

References 16 publications

Monolingual Phrase Alignment on Parse Forests

Monolingual Phrase Alignment on Parse Forests

Constrained Partial Parsing Based Dependency Tree Projection for Tree-to-Tree Machine Translation

The Study of the Effect of Educational Level of Students and English Language Teachers’ Awareness on True and False Cognates in Iran

Contact Info

Product

Resources

About