TY - JOUR
T1 - Egyptian Arabic to English Statistical Machine Translation System for NIST OpenMT'2015
AU - Sajjad, Hassan
AU - Durrani, Nadir
AU - Guzman, Francisco
AU - Nakov, Preslav
AU - Abdelali, Ahmed
AU - Vogel, Stephan
AU - Salloum, Wael
AU - El Kholy, Ahmed
AU - Habash, Nizar
PY - 2015/6/18
Y1 - 2015/6/18
N2 - The paper describes the Egyptian Arabic-to-English statistical machine translation (SMT) system that the QCRI-Columbia-NYUAD (QCN) group submitted to the NIST OpenMT'2015 competition. The competition focused on informal dialectal Arabic, as used in SMS, chat, and speech. Thus, our efforts focused on processing and standardizing Arabic, e.g., using tools such as 3arrib and MADAMIRA. We further trained a phrase-based SMT system using state-of-the-art features and components such as operation sequence model, class-based language model, sparse features, neural network joint model, genre-based hierarchically-interpolated language model, unsupervised transliteration mining, phrase-table merging, and hypothesis combination. Our system ranked second on all three genres.
AB - The paper describes the Egyptian Arabic-to-English statistical machine translation (SMT) system that the QCRI-Columbia-NYUAD (QCN) group submitted to the NIST OpenMT'2015 competition. The competition focused on informal dialectal Arabic, as used in SMS, chat, and speech. Thus, our efforts focused on processing and standardizing Arabic, e.g., using tools such as 3arrib and MADAMIRA. We further trained a phrase-based SMT system using state-of-the-art features and components such as operation sequence model, class-based language model, sparse features, neural network joint model, genre-based hierarchically-interpolated language model, unsupervised transliteration mining, phrase-table merging, and hypothesis combination. Our system ranked second on all three genres.
M3 - Article
JO - NIST OpenMT15 Evaluation
JF - NIST OpenMT15 Evaluation
ER -