Statistical Models for Unsupervised, Semi-Supervised, and Supervised Transliteration Mining

• Published in: Computational linguistics - Association for Computational Linguistics 2017-06, Vol.43 (2), p.349-375
• Main Authors: Sajjad, Hassan, Schmid, Helmut, Fraser, Alexander, Schütze, Hinrich
• Format: Article
• Language: English
• Subjects: Algorithms; Data mining; Natural language generation; Noise generation; Parallel corpora; Statistical models; Statistics; Training; Translations; Transliteration
• ISSN: 0891-2017; 1530-9312
• DOI: 10.1162/COLI_a_00286

We present a generative model that efficiently mines transliteration pairs in a consistent fashion in three different settings: unsupervised, semi-supervised, and supervised transliteration mining. The model interpolates two sub-models, one for the generation of transliteration pairs and one for the generation of non-transliteration pairs (i.e., noise). The model is trained on noisy unlabeled data using the EM algorithm. During training the transliteration sub-model learns to generate transliteration pairs and the fixed non-transliteration model generates the noise pairs. After training, the unlabeled data is disambiguated based on the posterior probabilities of the two sub-models. We evaluate our transliteration mining system on data from a transliteration mining shared task and on parallel corpora. For three out of four language pairs, our system outperforms all semi-supervised and supervised systems that participated in the NEWS 2010 shared task. On word pairs extracted from parallel corpora with fewer than 2% transliteration pairs, our system achieves up to 86.7% F-measure with 77.9% precision and 97.8% recall.