Improved Arabic–Chinese Machine Translation with Linguistic Input Features

Fares Aqlan,Xiaoping Fan,Abdullah Alqwbani,Akram Al-Mansoub

doi:10.3390/fi11010022

Abstract

This study presents linguistically augmented models of phrase-based statistical machine translation (PBSMT) using different linguistic features (factors) on the top of the source surface form. The architecture addresses two major problems occurring in machine translation, namely the poor performance of direct translation from a highly-inflected and morphologically complex language into morphologically poor languages, and the data sparseness issue, which becomes a significant challenge under low-resource conditions. We use three factors (lemma, part-of-speech tags, and morphological features) to enrich the input side with additional information to improve the quality of direct translation from Arabic to Chinese, considering the importance and global presence of this language pair as well as the limitation of work on machine translation between these two languages. In an effort to deal with the issue of the out of vocabulary (OOV) words and missing words, we propose the best combination of factors and models based on alternative paths. The proposed models were compared with the standard PBSMT model which represents the baseline of this work, and two enhanced approaches tokenized by a state-of-the-art external tool that has been proven to be useful for Arabic as a morphologically rich and complex language. The experiment was performed with a Moses decoder on freely available data extracted from a multilingual corpus from United Nation documents (MultiUN). Results of a preliminary evaluation in terms of BLEU scores show that the use of linguistic features on the Arabic side considerably outperforms baseline and tokenized approaches, the system can consistently reduce the OOV rate as well.

Highlights

With the rapid development of communication technology and economic globalization, translation between languages has become increasingly frequent, thereby drawing growing attention to machine translation (MT)
While neural machine translation (NMT) shows state-of-the-art performance on the plentiful training data, phrase-based Statistical MT (SMT) (PBSMT) still competitive or superior in the case of the low resource we focus on [15]
The authors in [25] discussed the benefits of enriching the input with morphological features to enhance the translation from English into Hindi and Marathi; the results showed that the integrated models reduced the number of out of vocabulary (OOV) words and improved the fluency of the translation

Summary

Introduction

With the rapid development of communication technology and economic globalization, translation between languages has become increasingly frequent, thereby drawing growing attention to machine translation (MT). The results showed that transfer learning provides an efficient performance in most cases, for transferring from multiple operating conditions to single operating conditions, the transfer learning led to a worse result Another line of work in low-resource scenarios is speech-to-text translation (ST) that has been proved valuable, for instance in the documentation of unwritten or endangered languages. Due to the scarcity of available documentation scenario that matches Arabic and Chinese as well as the expensive process and the required knowledge of domain to obtain hand-labeled training data, we leave the investigation of ASR technology between this language pair for future work, since this area is bringing more questions than answers.

Related Works

Challenges and Approach

Chinese Segmentation

Phrase-Based MT Models

Results and Analysis