One of the existing difficulties in natural lan­ guage processing applications is the lack of appropri­ ate tools for the recognition, translation, and/or translit­ eration of named entities (NEs), specifically for lessresourced languages. In this paper, we propose a new method to automatically label multilingual parallel data for Arabic-French pair of languages with named entity tags and build...

Statistical machine translation is quite robust when it comes to the choice of input representation. It only requires consistency between training and testing. As a result, there is a wide range of possible preprocessing choices for data used in statistical machine translation. This is even more so for morphologically rich languages such as Arabic. In this paper, we study the effect of differen...

This article aims to investigate and evaluate the translation of verb-noun collocation in English into Arabic Google and Bing online translation engines. A number of sentences were used as a testing dataset to evaluate both engines. Human translations by three bilingual speakers were used as a gold standard. A simple evaluation metric was proposed to calculate the translation accuracy of verb-n...

In this paper, we describe two approaches to extending syntactic constraints in the Hierarchical Phrase-Based (HPB) Statistical Machine Translation (SMT) model using Combinatory Categorial Grammar (CCG). These extensions target the limitations of previous syntax-augmented HPB SMT systems which limit the coverage of the syntactic constraints applied. We present experiments on Arabic–English and ...

We describe the TÜBITAK-UEKAE system that participated in the Arabic-to-English and Japanese-toEnglish translation tasks of the IWSLT 2007 evaluation campaign. Our system is built on the open-source phrasebased statistical machine translation software Moses. Among available corpora and linguistic resources, only the supplied training data and an Arabic morphological analyzer are used in the sys...

A Machine Translation (MT) system is an automatic process that translates from one human language to another language by using context information. We evaluate the use of an MT-based approach for query translation in an Arabic-English Cross-Language Information Retrieval (CLIR) system. We empirically evaluate the use of an MT-based approach for query translation in an Arabic-English CLIR system...

This paper describes the MIT-LL/AFRL statistical MT system and the improvements that were developed during the IWSLT 2009 evaluation campaign. As part of these efforts, we experimented with a number of extensions to the standard phrase-based model that improve performance on the Arabic and Turkish to English translation tasks. We discuss the architecture of the MIT-LL/AFRL MT system, improvemen...

The paper presents a research in Arabic Information Retrieval (IR). It surveys the impact of statistical and morphological analysis of Arabic text in improving Arabic IR relevancy. We investigated the contributions of Stemming, Indexing, Query Expansion, Text Summarization (TS), Text Translation, and Named Entity Recognition (NER) in enhancing the relevancy of Arabic IR. Our survey emphasizing ...

Automatic processing of mathematical information on the web imposes some difficulties. This paper presents a novel technique for automatic generation of mathematical equations semantic and Arabic translation on the web. The proposed system facilitates unambiguous representation of mathematical equations by correlating equations to their known names. The ability to extract the equation meaning f...

We present in this paper PADIC, a Parallel Arabic DIalect Corpus we built from scratch, then we conducted experiments on crossdialect Arabic machine translation. PADIC is composed of dialects from both the Maghreb and the Middle-East. Each dialect has been aligned with Modern Standard Arabic (MSA). Three dialects from Maghreb are concerned by this study: two from Algeria, one from Tunisia, and ...