DALILA: The Dialectal Arabic Linguistic Learning Assistant

DALILA: The Dialectal Arabic Linguistic Learning Assistant Salam Khalifa, Houda Bouamor†, Nizar Habash Computational Approaches to Modeling Language Lab, New York University Abu Dhabi, UAE †Carnegie Mellon University in Qatar, Qatar {salamkhalifa,nizar.habash}@nyu.edu, hbouamor@cmu.edu Abstract Dialectal Arabic (DA) poses serious challenges for Natural Language Processing (NLP). The number and sophistication of tools and datasets in DA are very limited in comparison to Modern Standard Arabic (MSA) and other languages. MSA tools do not effectively model DA which makes the direct use of MSA NLP tools for handling dialects impractical. This is particularly a challenge for the creation of tools to support learning Arabic as a living language on the web, where authentic material can be found in both MSA and DA. In this paper, we present the Dialectal Arabic Linguistic Learning Assistant (DALILA), a Chrome extension that utilizes cutting-edge Arabic dialect NLP research to assist learners and non-native speakers in understanding text written in either MSA or DA. DALILA provides dialectal word analysis and English gloss corresponding to each word. Keywords: Dialectal Arabic, Computer Assisted Language Learning, Morphology 1. Introduction Computer-assisted language learning is an area where re- searchers study how to make use of computers to support learners of a certain language (Chapelle, 2008; Levy and Stockwell, 2013). Many software tools were designed to help learners by providing dictionaries, expert feedback, carefully designed “immersive” experiences, etc. However, language learners see little support online especially with real raw materials in the language they are learning. There is no substitute for a language learner to go online and read and understand text written by native speakers. In the case of Arabic, the situation is more challenging than English and other European languages. The prob- lem that most learners of Arabic face is that the written language is radically different from the various forms of Dialectal Arabic (DA) spoken throughout the Arab World. Hence, even the most advanced learners of Modern Stan- dard Arabic (MSA) or Moroccan Arabic (to pick a dialect randomly) will find themselves confused on the streets of Cairo as well as on microblogging websites. It is also im- portant to note that even "native" speakers have trouble un- derstanding other dialect vocabulary (Holes, 2004). Differ- ent DAs differ phonologically, lexically, morphologically, and syntactically from one another and from MSA (Holes, 1986; Watson, 2007; Brustad, 2000; Habash and Rambow, 2006; Habash et al., 2012; Bouamor et al., 2014). For ex- ample, state-of-the-art MSA morphological analyzers have been shown to have only 60% coverage of Levantine Ara- bic verbs (Habash and Rambow, 2006) and 64% coverage of Egyptian Arabic words (Habash et al., 2012). While MSA is commonly used in formal written contexts, people are increasingly using DA on social media platforms such as Facebook, Twitter or Youtube to share their sto- ries, opinions, post a comment or give a feedback or com- ment on a video and interact with the community (Salama et al., 2014; Sadat et al., 2014). In a recent analysis, Huang (2015) showed that 42% of a …

Content truncated to 500 words