A Leveled Reading Corpus of Modern Standard Arabic

A Leveled Reading Corpus of Modern Standard Arabic Muhamed Al Khalil,∗ Hind Saddiki,∗† Nizar Habash,∗ Latifa Alfalasi‡ ∗New York University Abu Dhabi, UAE †Mohammed V University in Rabat, Morocco ‡Ministry of Education, UAE {muhamed.alkhalil, hind.saddiki, nizar.habash}@nyu.edu, latifa.alfalasi@moe.gov.ae Abstract We present a reading corpus in Modern Standard Arabic to enrich the sparse collection of resources that can be leveraged for educational applications. The corpus consists of textbook material from the curriculum of the United Arab Emirates, spanning all 12 grades (1.4 million tokens) and a collection of 129 unabridged works of fiction (5.6 million tokens) all annotated with reading levels from Grade 1 to Post-secondary. We examine reading progression in terms of lexical coverage, and compare the two sub-corpora (curricular, fiction) to others from clearly established genres (news, legal/diplomatic) to measure representation of their respective genres. Keywords: Arabic, Corpus, Leveled Reading, Curriculum, Fiction 1. Introduction Corpora are built for a wide range of purposes such as mod- eling language use for linguistics research, instructional material for educators, or training data for natural language processing (NLP) applications. Continued efforts in creat- ing such resources are instrumental in furthering research for all application domains of NLP, namely, parsing and part-of-speech (POS) tagging, speech recognition, machine translation, document classification, etc. Work in NLP for Modern Standard Arabic (MSA) is gain- ing momentum as more resources and tools are developed (Habash, 2010). Corpus data for MSA has been mostly sourced from the news genre (Zaghouani, 2014), while there are far fewer specialized resources, such as corpora for educational applications (Zaghouani et al., 2014; Al- faifi et al., 2013). As a particular type of educational re- source, a level-annotated reading corpus can be leveraged for a multitude of applications: text simplification, auto- matic readability assessment, computer-assisted language learning, data-driven pedagogy, text genre and register pro- filing, and so on. Building a corpus of this nature con- tributes to the variety of resources at our disposal, allowing for research in Arabic NLP to progress in new directions. In this paper, we present a reading corpus in MSA collected from textbooks of the United Arab Emirates (UAE) curricu- lum and a collection of 129 unabridged works of fiction. The curriculum texts are labeled with levels from grade 1 to 12 and the fiction texts are at a Post-secondary level, i.e., adult-level reading that is accessible to someone after achieving 12th grade reading proficiency. This corpus was created in the context of a project on the Simplification of Arabic Masterpieces for Extensive Reading (SAMER) in- tended to simplify works of Arabic fiction to a level that is more accessible for school-aged readers (Al Khalil et al., 2017). The paper is organized as follows. Section 2 presents re- lated work in corpus creation; Section 3 describes the cor- pus collection and annotation; We analyze the data in Sec- tion 4 before stating our conclusions and future work. 2. Background and Related Work The multi-faceted complexity of MSA makes it a challeng- ing language to tackle in NLP. There is the …

Content truncated to 500 words