Extracting Parallel Sentences with Bidirectional Recurrent Neural Networks to Improve Machine Translation
arXiv:1806.05559
Abstract
Parallel sentence extraction is a task addressing the data sparsity problem found in multilingual natural language processing applications. We propose a bidirectional recurrent neural network based approach to extract parallel sentences from collections of multilingual texts. Our experiments with noisy parallel corpora show that we can achieve promising results against a competitive baseline by removing the need of specific feature engineering or additional external resources. To justify the utility of our approach, we extract sentence pairs from Wikipedia articles to train machine translation systems and show significant improvements in translation performance.
arXiv comment12 pages, 7 figures, COLING 2018. arXiv admin note: text overlap with arXiv:1709.09783
References in corpus (9)
- Adam: A Method for Stochastic Optimization
- Neural Machine Translation by Jointly Learning to Align and Translate
- Sequence to Sequence Learning with Neural Networks
- Recurrent Neural Network Regularization
- Dual Learning for Machine Translation
- MINE: Mutual Information Neural Estimation
- BilBOWA: Fast Bilingual Distributed Representations without Word Alignments
- Word Translation Without Parallel Data
- Unsupervised Neural Machine Translation