QuEst – Design, Implementation and Extensions of a Framework for Machine Translation Quality Estimation

Shah, Kashif; Avramidis, Eleftherios; Biçici, Ergun; Specia, Lucia

doi:10.2478/pralin-2013-0008

Cited by 19 publications

(34 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…RTM PPP models are built for each task emerging from training set, test set, and label set obtained from a parser. RTMs achieve top results in machine translation performance prediction (MTPP) in quality estimation task (Biçici et al, 2015b;Biçici, 2016), can achieve better results than open-source MTPP tool QuEst (Shah et al, 2013;Biçici and Specia, 2015), and can achieve top results in semantic similarity prediction tasks . We provide a current picture on PPP detailing prediction performance, top selected features, and lower bound on prediction error of PPP.…”

Section: Predicting Parsing Performance With Referential Translation mentioning

confidence: 99%

Predicting the Performance of Parsing with Referential Translation Machines

Biçici

2016

The Prague Bulletin of Mathematical Linguistics

Self Cite

View full text Add to dashboard Cite

Referential translation machine (RTM) is a prediction engine used for predicting the performance of natural language processing tasks including parsing, machine translation, and semantic similarity pioneering language, task, and domain independence. RTM results for predicting the performance of parsing (PPP) in out-of-domain or in-domain settings with different training sets and types of features present results independent of language or parser. RTM PPP models can be used without parsing using only text input and without any parser or language dependent information. Our results detail prediction performance, top selected features, and lower bound on the prediction error of PPP.

show abstract

Section: Predicting Parsing Performance With Referential Translation mentioning

confidence: 99%

Predicting the Performance of Parsing with Referential Translation Machines

Biçici

2016

The Prague Bulletin of Mathematical Linguistics

Self Cite

View full text Add to dashboard Cite

show abstract

“…The list of features and the way they are extracted were identical as described in [12]. More details could be found in [12,13,14].…”

Section: Features For Quality Estimationmentioning

confidence: 99%

Groupwise learning for ASR k-best list reranking in spoken language translation

Shah

Specia

et al. 2016

2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

This paper studies the enhancement of spoken language translation (SLT) with groupwise learning. Groupwise features were constructed by grouping pairs, triplets or M -plets of the ASR k-best outputs. Regression and classification models were learnt and a straightforward score combination strategy was used to capture the ranking relationship. Groupwise learning with pairwise regression models give the biggest gain over simple support vector regression models. Groupwise learning is robust to sentences with different ASR-confidence, meaning that the confidence threshold heuristics in reranking are no longer needed. This technique is also complementary to linear discriminant analysis feature projection. Altogether a BLEU score improvement of 0.80 was achieved in an indomain English-to-French SLT task.

show abstract

“…Qualitative (Avramidis, 2016) is used for the feature generation and prediction of sentence-level ranking, whereas models produced by the popular state-ofthe-art QuEST (Shah et al, 2013) can be used for continuous score prediction. The database structure allows for loading and storing other Python-based QE models, provided that their usage is clearly documented.…”

Section: Quality Estimationmentioning

confidence: 99%

QE::GUI – A Graphical User Interface for Quality Estimation

Avramidis¹

2017

The Prague Bulletin of Mathematical Linguistics

Self Cite

View full text Add to dashboard Cite

Despite its wide applicability, Quality Estimation (QE) of Machine Translation (MT) poses a difficult entry barrier since there are no open source tools with a graphical user interface (GUI). Here we present a tool in this direction by connecting the back-end of the QE decision-making mechanism with a web-based GUI. The interface allows the user to post requests to the QE engine and get a visual response with the results. Additionally we provide pre-trained QE models for easier launching of the app. The tool is written in Python so that it can leverage the rich natural language processing capabilities of the popular dynamic programming language, which is at the same time supported by top web-server environments.

show abstract

QuEst – Design, Implementation and Extensions of a Framework for Machine Translation Quality Estimation

Cited by 19 publications

References 6 publications

Predicting the Performance of Parsing with Referential Translation Machines

Predicting the Performance of Parsing with Referential Translation Machines

Groupwise learning for ASR k-best list reranking in spoken language translation

QE::GUI – A Graphical User Interface for Quality Estimation

Contact Info

Product

Resources

About