Nizar Habash · page 5
Papers
- ArzEn-ST: A Three-way Speech Translation Corpus for Code-Switched Egyptian Arabic-English (2022 · 26 citations)
Code-switching, where speakers mix languages within speech, is common in multilingual communities. NLP systems still handle it poorly, mostly because there is little data. This paper introduces ArzEn-ST, a speech translation resource for Egyptian Arabic mixed with English. It builds on the ArzEn corpus, which… - A Leveled Reading Corpus of Modern Standard Arabic (2018 · 26 citations)
Most corpora for Modern Standard Arabic (MSA) come from news, so educational resources are scarce. This matters especially for readability assessment, text simplification and language learning. The authors built a reading corpus with level labels and two parts. The first part is the complete UAE school curriculum… - Using Deep Morphology to Improve Automatic Error Detection in Arabic Handwriting Recognition (2011 · 26 citations)
Habash and Roth address a sub-problem of Arabic handwriting recognition (HR). Instead of fixing recognition mistakes, they only try to locate them. Arabic is hard for HR because letters connect and diacritic dots are small or ambiguous. Its rich morphology also makes word-list lookup unreliable. The authors ran the… - Morphological Analysis and Disambiguation for Gulf Arabic: The Interplay between Resources and Methods (2020 · 25 citations)
This is the first full morphological analysis and disambiguation system for Gulf Arabic (Emirati data). The analyser for Gulf Arabic is not written by hand: it is built by paradigm completion from the training split of the Annotated Gumar Corpus (about 202,000 tokens, eight novel excerpts). Two hand-built analysers… - Building an Arabic Machine Translation Post-Edited Corpus: Guidelines and Annotation (2016 · 24 citations)
Automatic post-editing of machine translation needs training data: MT output paired with human corrections. - MedArabiQ: Benchmarking Large Language Models on Arabic Medical Tasks (2025 · 22 citations)
LLMs are increasingly considered for clinical use, but there has been almost no way to check how they handle medical content in Arabic. - The Effects of Factorizing Root and Pattern Mapping in Bidirectional Tunisian - Standard Arabic Machine Translation (2013 · 22 citations)
Arabic dialects such as Tunisian (TUN) have almost no NLP resources, while Modern Standard Arabic (MSA) has many. - The Arabic Parallel Gender Corpus 2.0: Extensions and Analyses (2022 · 21 citations)
NLP systems such as machine translation usually produce a single output with no knowledge of who is speaking or who is being addressed. - Improving Arabic Dependency Parsing with Form-based and Functional Morphological Features (2011 · 21 citations)
The authors ask which morphological features help a transition-based parser (MaltParser, eager algorithm, SVM classifiers) on Modern Standard Arabic. - An Arabic Dependency Treebank in the Travel Domain (2019 · 20 citations)
The authors build a small dependency treebank of Modern Standard Arabic from the travel-phrase corpus BTEC, in the Columbia Arabic Treebank (CATiB) style, so that it can be compared with the news-based Penn Arabic Treebank. - POS-tagging of Tunisian Dialect Using Standard Arabic Resources and Tools (2015 · 20 citations)
Tunisian Arabic (TUN) is a spoken dialect with about twelve million speakers, and it has almost no annotated data. - Generation from Lexical Conceptual Structures (2000 · 20 citations)
The paper describes a generator that turns Lexical Conceptual Structures (LCS), a compositional, language-independent interlingua, into English sentences. - A Unified Model for Arabizi Detection and Transliteration using Sequence-to-Sequence Models (2020 · 19 citations)
Egyptian Arabic on social media is often typed in Roman letters (Arabizi). This text is noisy, has no fixed spelling, and mixes freely with English. Earlier systems treated two jobs separately: deciding which words are Arabizi, and converting them to Arabic script. This paper handles both with one character-level… - Semi-automatic error analysis for large-scale statistical machine translation (2007 · 19 citations)
The paper relates properties of the input documents to the translation quality of a large SMT system, scored per document by HTER. - BAREC Shared Task 2025 on Arabic Readability Assessment (2025 · 18 citations)
This report covers the first shared task on fine-grained readability assessment for Arabic, held at ArabicNLP 2025. - Guidelines for Fine-grained Sentence-level Arabic Readability Annotation (2025 · 18 citations)
The paper addresses the lack of a large, fine-grained, manually labeled resource for judging how hard Arabic text is to read. - Complementary Strategies for Low Resourced Morphological Modeling (2018 · 18 citations)
Morphologically rich languages spread each lexeme across many inflected forms. Individual word types therefore occur rarely, and models get little contextual evidence for each one. The authors ask which of two common remedies is more worthwhile for a low-resourced language: hand-written, context-free morphological… - Camel Treebank: An Open Multi-genre Arabic Dependency Treebank (2022 · 17 citations)
Camel Treebank 1.0 is an open dependency treebank of about 188k words (about 242k tokens) of Modern Standard and Classical Arabic, in the Columbia Arabic Treebank (CATiB) style. - Hierarchical Aggregation of Dialectal Data for Arabic Dialect Identification (2022 · 17 citations)
Arabic dialect identification (DID) resources have grown quickly, but each corpus labels its texts differently. - Machine Translation between Hebrew and Arabic: Needs, Challenges and Preliminary Solutions (2010 · 17 citations)
Modern Hebrew and Standard Arabic share orthography, root-and-pattern morphology, clitics and much of their vocabulary, but they differ in case, dual number, broken plurals, the feminine ending in genitive constructions and the agreement of irrational plural nouns. - Exploiting Arabic Diacritization for High Quality Automatic Annotation (2016 · 16 citations)
Undiacritized Arabic is highly ambiguous: about 12.8 analyses per word out of context. The authors use the diacritics that a text does have to choose among those analyses. MADAMIRA (with the SAMA analyzer) ranks the candidates as usual; then the candidates are filtered by the edit distance between their diacritized… - Identifying Broken Plurals, Irregular Gender, and Rationality in Arabic Text (2012 · 16 citations)
Arabic words can carry a mismatch between form and function: a broken plural looks singular, and some nouns have a feminine form with masculine agreement, or the reverse. - NADI 2025: The First Multidialectal Arabic Speech Processing Shared Task (2025 · 15 citations)
NADI 2025 is the sixth edition of the Nuanced Arabic Dialect Identification series. Earlier editions worked on written text. This one moves to speech, because everyday Arabic is mostly spoken dialect, and few datasets annotate it. The shared task has three linked subtasks: - Hebrew Morphological Preprocessing for Statistical Machine Translation (2012 · 15 citations)
The authors compare ways of splitting Hebrew words into tokens before Hebrew-to-English statistical translation with Moses, on a training set of about 64,000 sentences. - Morphotactic Modeling in an Open-source Multi-dialectal Arabic Morphological Analyzer and Generator (2022 · 14 citations)
The paper describes CAMeL Morph, an open framework for writing Arabic morphology as specifications and compiling them into a database that the Camel Tools analyser and generator can use.
This page in Markdown → https://md.research.elamur.ai/free/en/authors/nizar-habash/page/5/