Neural Arabic Text Diacritization: State of the Art Results and a Novel Approach for Machine Translation
This work addresses the problem of automatic diacritization for Arabic text, which is incremental as it builds on existing deep learning methods with specific enhancements.
The authors tackled Arabic text diacritization using deep learning models like FFNN and RNN with enhancements such as 100-hot encoding and CRF, achieving results on par or better than existing models without language-dependent post-processing. They also proposed a Translation over Diacritization approach to enhance machine translation models using diacritics.
In this work, we present several deep learning models for the automatic diacritization of Arabic text. Our models are built using two main approaches, viz. Feed-Forward Neural Network (FFNN) and Recurrent Neural Network (RNN), with several enhancements such as 100-hot encoding, embeddings, Conditional Random Field (CRF) and Block-Normalized Gradient (BNG). The models are tested on the only freely available benchmark dataset and the results show that our models are either better or on par with other models, which require language-dependent post-processing steps, unlike ours. Moreover, we show that diacritics in Arabic can be used to enhance the models of NLP tasks such as Machine Translation (MT) by proposing the Translation over Diacritization (ToD) approach.