Segmentation for English-to-Arabic Statistical Machine Translation

By Ibrahim Badr, Rabih Zbib, James R. Glass. 2008

81 citations recorded by the source corpus.

In this paper, we report on a set of initial results for English-to-Arabic Statistical Machine Translation (SMT). We show that morphological decomposition of the Arabic source is beneficial, especially for smaller-size corpora, and investigate different recombination techniques. We also report on the use of Factored Translation Models for English-to-Arabic translation.

Subfields: Artificial Intelligence, Molecular Biology, Machine Translation

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Molecular Biology

In Machine Translation

Citations in the corpus

What do Neural Machine Translation Models Learn about Morphology? (2017, methodology): “…systems that deal with morphologically-rich languages resort to various techniques for representing morphological knowledge, such as word segmentation (Nieﬂen and Ney, 2000 Knight, 2003; Badr et al., 2008) and factored translation and reordering models (Koehn and Hoang, 2007; Durrani et al., 2014).”
MADA + TOKAN : A Toolkit for Arabic Tokenization , Diacritization , Morphological Disambiguation , POS Tagging , Stemming and Lemmatization (2009, methodology): “Combination of Preprocessing Schemes for Statistical MT.”
A Hybrid Morpheme-Word Representation for Machine Translation of Morphologically Rich Languages (2010, background): “These approaches, however, either only succeed in enhancing the performance for small bi-texts (Badr et al., 2008; Oflazer and El-Kahlout, 2007), or improve only modestly for large bi-texts1.”
Rich Morphology Generation Using Statistical Machine Translation (2012, background): “We follow numerous previously published efforts on the value of tokenization for Arabic NLP tasks (Badr et al., 2008; El Kholy and Habash, 2010).”
On internal language representations in deep learning: an analysis of machine translation and speech recognition (2018, background): “Super-word levels texts utterances sentences phrases relations words morphems syllables characters phonemes Sub-word levels been shown to improve machine translation performance [6, 15, 139, 184, 258].”
Large-Scale Machine Translation between Arabic and Hebrew: Available Corpora and Initial Results (2016, background): “In particular, the importance of morphology-aware tokenization when translating from and to Arabic has been conﬁrmed in phrase-based (Badr et al., 2008; Habash and Sadat, 2006; El Kholy and Habash, 2012) and neural machine translation, in both hybrid (Devlin et al., 2014) and end-to-end systems…”
Unsupervised Morphology Rivals Supervised Morphology for Arabic MT (2012, methodology): “The approach we have taken in this paper is an instance of a segmented MT model, which divides the input into morphemes and uses the derived morphemes as a unit of translation (Sadat and Habash, 2006; Badr et al., 2008; Clifton and Sarkar, 2011).”
Book Reviews: Introduction to Arabic Natural Language Processing by Nizar Y. Habash (2010, methodology): “The following is a description of some commonly used tokenization schemes [83, 105, 107, 99, 13].”
Syntactic Phrase Reordering for English-to-Arabic Statistical Machine Translation (2009, methodology): “The only works on English-to-Arabic SMT that we are aware of are Badr et al. (2008), and Sarikaya and Deng (2007).”