About
I am a PhD candidate and professional lecturer at The George Washington University. I work with Prof. Mona Diab on natural language processing research, particularly vector-space models for words and sentences, cross-lingual mapping of such models, and their applications in paraphrasing and machine translation.
Projects
Currently, I am working on learning general-purpose cross-lingual vector representation that can be used for semantic textual similarity, document classification, paraphrasing, and machine translation using matrix factorization and neural networks. I am also investigating unsupervised methods for deducing dictionaries from independent vector-space models. Previously, I worked on Arabic part-of-speech tagging as part of the MADAMIRA project.
Teaching
NLP Publications
Hanan Aldarmaki and Mona Diab. "Evaluation of Unsupervised Compositional Representations." In Processings of The 27th International Conference on Computational Linguistics (COLING), 2018. [pdf]
Hanan Aldarmaki, Mahesh Mohan, and Mona Diab. "Unsupervised Word Mapping Using Structural Similarities in Monolingual Embeddings." Transactions of the Association for Computational Linguistics (TACL), 2018. [pdf]
Hanan Aldarmaki and Mona Diab. "Learning Cross-lingual Representations with Matrix Factorization." Workshop on Multilingual and Cross-lingual Methods in NLP, 2016. [pdf]
Hanan Aldarmaki and Mona Diab. "GWU NLP at SemEval-2016 Shared Task 1: Matrix Factorization for Crosslingual STS." Proceedings of SemEval, 2016. [pdf]
Robust Part-of-Speech Tagging of Arabic Text. Hanan Aldarmaki & Mona Diab. Proceedings of the second workshop on Arabic Natural Language Processing (ANLP 2015). [pdf]
Learn More
Email me here if you have questions.