Christian Khairallah (Cayralat)
كريستيان خيرالله
كريستيان خيرالله
I am currently embarking on the entrepreneurship journey, and on my way to building my product, Aralects, the Arabic Dialect language learning app. Before that, I freelanced as a Speech Technology Scientist as part of the Machine Translation team at AppTek. Additionally, I am affiliated with New York University Abu Dhabi as a visiting scholar [page], where I am finishing up papers, and consulting on a few matters.
Just before that, I was a full-time research assistant at the Computational Approaches to Modeling Language (CAMeL) Lab at New York University Abu Dhabi, under the supervision of Prof. Nizar Habash.
I graduated with distinction in Electrical and Computer Engineering from the American University of Beirut (AUB) with a minor in English Language, and with a double Master of Science in Computational Linguistics from Charles University in Prague and Saarland University in Germany, where I attended both as part of the Language and Communication Technologies (LCT) Erasmus Mundus Master's program.
During my Master's, I focused on processing spontaneous orthography in Dialectal Arabic, which is the result of it lacking any standard orthography, working on tasks such as morphological analysis and segmentation, character-level neural machine translation, spelling correction, and taxonomy and dataset creation.
On the personal level, I am a big language enthusiast, and I am currently especially interested in the history of Semitic languages, and more specifically in the Arabic branch and how current variants came to co-exist with Classical Arabic in a diglossic relationship. I am natively fluent in Arabic, French, and English, beginner-level in German, and I am currently learning Italian. In my spare time, I am an avid mélomane and I am interested in the history of music in general, I mostly go hiking, cycle, swim, play ping pong, practice yoga, cook, and father a crew of insufferable indoor plants.
NYUAD
In my work, I focused on computational approaches aiming to reconcile the disparity between Modern Standard Arabic and Dialectal Arabic, both in terms of resource creation and processing tools. In addition to leveraging the latest computational methods to solve current NLP problems, I took a keen interest in dialectal and standard Arabic morphology and syntax over the span of two years. Here are the things I worked on (or am still involved in):
Working on maintaining the Maknuune Palestinian Arabic Lexicon. Check out the PDF Book version that I created for it!
Camel Morph Project
Maintained the Camel Morph repository which is a project aiming to build large open-source morphological models for Arabic and its dialects across many genres and domains [github].
Co-supervised final-year capstone students on extracting dialectal Arabic lexicons in a semi-supervised way.
Worked on morphological analyzers/generators for Modern Standard Arabic, Egyptian, Gulf, Levantine and Tunisian Arabic.
Worked on updating the Conventional Orthography for Dialectal Arabic (CODA*) guidelines [paper]
Publications
Camel Morph MSA: A Large-Scale Open-Source Morphological Analyzer for Modern Standard Arabic
Christian Khairallah, Salam Khalifa, Reham Marzouk, Mayar Nassar, Nizar Habash
In Proceedings of the Joint International Conference on Computational Linguistics, Language Resources and Evaluation, LREC-COLING (Turin, 2024)
[paper] [github]Computational Morphology and Lexicography Modeling of Modern Standard Arabic Nominals
Christian Khairallah, Reham Marzouk, Salam Khalifa, Mayar Nassar, Nizar Habash
In Findings of the 18th Conference of the European Chapter of the Association for Computational Linguistics, EACL (Malta, 2024)
[paper] [github]Advancements in Arabic Grammatical Error Detection and Correction: An Empirical Investigation
Bashar Alhafni, Go Inoue, Christian Khairallah, Nizar Habash
In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, EMNLP (Singapore, 2023)
[paper]Maknuune: A Large Open Palestinian Arabic Lexicon
Shahd Dibas, Christian Khairallah, Nizar Habash, Omar Fayez Sadi, Tariq Sairafy, Karmel Sarabta, Abrar Ardah
In Proceedings of the Sixth Arabic Natural Language Processing Workshop co-located with EMNLP (Abu Dhabi, 2022)
[paper] [website] [pdf book]Morphotactic Modeling in an Open-source Multi-dialectal Arabic Morphological Analyzer and Generator
Nizar Habash, Reham Marzouk, Christian Khairallah, Salam Khalifa
In Proceedings of the Nineteenth SIGMORPHON Workshop co-located with NAACL (Seattle, 2022)
[paper] [github]Orthography Standardization in Arabic Dialects
Christian Khairallah
Master's Thesis (Prague & Saarbrücken, 2021)
[report] [code] [data]