Using Word Embeddings for Improving Statistical Machine Translation
            of Phrasal Verbs

Cholakov, Kostadin; Kordoni, Valia

doi:10.18653/v1/w16-1808

Cited by 6 publications

(4 citation statements)

References 9 publications

(12 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bilingual embeddings are vector representations of two languages mapped into shared space, such that translated word pairs have similar vectors (Gouws et al, 2015;. They facilitate applications from parallel sentence extraction (Grover and Mitra, 2017) to machine translation (Zou et al, 2013;Cholakov and Kordoni, 2016;Artetxe et al, 2017b) and can be used to improve monolingual embeddings (Faruqui and Dyer, 2014). Bilingual embeddings are learned via one of three methods: mapping both spaces into a shared space (Mikolov et al, 2013b), monolingual adaptation of one language's embedding space into another's (Zou et al, 2013), or bilingually training both embeddings simultaneously (AP et al, 2014;.…”

Section: Related Workmentioning

confidence: 99%

Addressing Noise in Multidialectal Word Embeddings

Erdmann¹,

Zalmout²,

Habash³

2018

Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)

View full text Add to dashboard Cite

Word embeddings are crucial to many natural language processing tasks. The quality of embeddings relies on large nonnoisy corpora. Arabic dialects lack large corpora and are noisy, being linguistically disparate with no standardized spelling. We make three contributions to address this noise. First, we describe simple but effective adaptations to word embedding tools to maximize the informative content leveraged in each training sentence. Second, we analyze methods for representing disparate dialects in one embedding space, either by mapping individual dialects into a shared space or learning a joint model of all dialects. Finally, we evaluate via dictionary induction, showing that two metrics not typically reported in the task enable us to analyze our contributions' effects on low and high frequency words. In addition to boosting performance between 2-53%, we specifically improve on noisy, low frequency forms without compromising accuracy on high frequency forms.

show abstract

Section: Related Workmentioning

confidence: 99%

Addressing Noise in Multidialectal Word Embeddings

Erdmann¹,

Zalmout²,

Habash³

2018

Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)

View full text Add to dashboard Cite

show abstract

“…Word embeddings have became commonly used in many MWErelated tasks due to their intriguing properties. For instance, word embeddings are user in order to improve the quality of the translation of phrasal verbs in [7].…”

Section: Previous Workmentioning

confidence: 99%

l1 Regularization of Word Embeddings for Multi-Word Expression Identification

Berend¹

2018

Acta Cybern

View full text Add to dashboard Cite

In this paper we compare the effects of applying various state-of-the-art word representation strategies in the task of multi-word expression (MWE) identification. In particular, we analyze the strengths and weaknesses of the usage of 1-regularized sparse word embeddings for identifying MWEs. Our earlier study demonstrated the effectiveness of regularized word embeddings in other sequence labeling tasks, i.e. part-of-speech tagging and named entity recognition, but it has not yet been rigorously evaluated for the identification of MWEs yet.

show abstract

“…It's due to this abundance of FSs in everyday written and spoken discourse that apart from traditional classroom interest in them, FSs are the main focus of interest in different fields, such as computer assisted language learning (CALL) (Stengers, et al, 2014), mobile learning (M-learning) (Hayati, et al, 2013), machine translation (Cholakov, et al, 2014), eye tracking (Siyanova-Chanturia, 2013; Siyanova-Chanturia, et al, 2011; Conklin, 2014, 2015), neurolinguistics (Boulenger, et al, 2012;Zhang, et al, 2013), and brain diseases/damages (Reuterskiöld, and Van Lancker-Sidtis, 2013; Van Lancker-Sidtis, 2006).…”

Section: Literature Reviewmentioning

confidence: 99%

The Effect of Two Attention-drawing Techniques on Learning English Idioms

Razmjoo¹,

Songhori²,

Bahremand³

2016

JLTR

View full text Add to dashboard Cite

Abstract-This study investigated the effectiveness of three methods for teaching idioms. Sixty two intermediate EFL learners from a language school in Kerman were divided into three classes to be taught in three conditions: the etymological elaboration experimental group (N=20) was taught 16 target idioms in a text along with the etymology of each idiom; the typographic salience experimental group (N=20) was taught the same text as the previous group but the 16 target idioms were in boldfaced type and red; and the traditional control group (N=22) was taught the same text in plain typeface with no etymology for idioms. The participants in the three groups were not informed about the posttest and the delayed posttest. The results of the comparison between the posttest and the delayed posttest scores of all the three groups indicate that the participants in etymological elaboration outperformed the participants in the other two groups in terms of idioms' retention and recall. These findings are suggestive that the most effective method of teaching idioms vis-a-vis their retention and recall is etymological elaboration.

show abstract

Using Word Embeddings for Improving Statistical Machine Translation of Phrasal Verbs

Cited by 6 publications

References 9 publications

Addressing Noise in Multidialectal Word Embeddings

Addressing Noise in Multidialectal Word Embeddings

l1 Regularization of Word Embeddings for Multi-Word Expression Identification

The Effect of Two Attention-drawing Techniques on Learning English Idioms

Contact Info

Product

Resources

About