Machine Translation with Minimal Reliance on Parallel Resources

Tambouratzis, George; Vassiliou, Marina; Sofianopoulos, Sokratis

doi:10.1007/978-3-319-63107-3

Cited by 4 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…PAM was developed as part of the PRESEMT hybrid MT methodology (Tambouratzis et al, 2017(Tambouratzis et al, 2017). PRESEMT was designed to create MT systems requiring only very limited amounts of specialized, expensive linguistic resources.…”

Section: Presemt Essentialsmentioning

confidence: 99%

Alignment verification to improve NMT translation towards highly inflectional languages with limited resources

Tambouratzis¹

2021

Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume

Self Cite

View full text Add to dashboard Cite

The present article studies translation quality when limited training data is available to translate towards morphologically rich languages. The starting point is a neural MT system, used to train translation models with only publicly available parallel data. An initial analysis of the translation output has shown that quality is sub-optimal, mainly due to the insufficient amount of training data. To improve translation, a hybridized solution is proposed, using an ensemble of relatively simple NMT systems trained with different metrics, combined with an open source module designed for low-resource MT that measures the alignment level. A quantitative analysis based on established metrics is complemented by a qualitative analysis of translation results. These show that over multiple test sets, the proposed hybridized method confers improvements over (i) both the best individual NMT and (ii) the ensemble system provided in the Marian-NMT package. Improvements over Marian-NMT are in many cases statistically significant.

show abstract

Section: Presemt Essentialsmentioning

confidence: 99%

Alignment verification to improve NMT translation towards highly inflectional languages with limited resources

Tambouratzis¹

2021

Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume

Self Cite

View full text Add to dashboard Cite

show abstract

“…The setup used is broadly similar to that of [6]. Our work uses a parallel bilingual corpus to extract a phrasing model, which is subsequently applied to the PRESEMT MT system (see [10] for a comprehensive description of PRESEMT). However, in contrast to Zhu et al [6], who employ a parallel corpus of 10,000 sentence pairs in English and Chinese, the corpus used in the present work consists of only a few hundred sentences in the source and target languages.…”

Section: Phrasing Model Outlinementioning

confidence: 99%

Swarm Algorithms for NLP - The Case of Limited Training Data

Tambouratzis

Vassiliou

2019

Journal of Artificial Intelligence and Soft Computing Research

Self Cite

View full text Add to dashboard Cite

The present article describes a novel phrasing model which can be used for segmenting sentences of unconstrained text into syntactically-defined phrases. This model is based on the notion of attraction and repulsion forces between adjacent words. Each of these forces is weighed appropriately by system parameters, the values of which are optimised via particle swarm optimisation. This approach is designed to be language-independent and is tested here for different languages. The phrasing model’s performance is assessed per se, by calculating the segmentation accuracy against a golden segmentation. Operational testing also involves integrating the model to a phrase-based Machine Translation (MT) system and measuring the translation quality when the phrasing model is used to segment input text into phrases. Experiments show that the performance of this approach is comparable to other leading segmentation methods and that it exceeds that of baseline systems.

show abstract

Selecting the Optimal Configuration ofSwarm Algorithms for an NLP Task

Tambouratzis¹

2021

Artificial Intelligence and Soft Computing

View full text Add to dashboard Cite

Machine Translation with Minimal Reliance on Parallel Resources

Cited by 4 publications

References 0 publications

Alignment verification to improve NMT translation towards highly inflectional languages with limited resources

Alignment verification to improve NMT translation towards highly inflectional languages with limited resources

Swarm Algorithms for NLP - The Case of Limited Training Data

Selecting the Optimal Configuration ofSwarm Algorithms for an NLP Task

Contact Info

Product

Resources

About