A Linguistic Evaluation of Rule-Based, Phrase-Based, and Neural MT Engines

Aljoscha Burchardt, Vivien Macketanz, Jon Dehdari, Georg Heigold, Jan-thorsten Peter, Philip Williams

Research output: Contribution to journalArticlepeer-review


In this paper, we report an analysis of the strengths and weaknesses of several Machine Translation (MT) engines implementing the three most widely used paradigms. The analysis is based on a manually built test suite that comprises a large range of linguistic phenomena. Two main observations are on the one hand the striking improvement of an commercial online system when turning from a phrase-based to a neural engine and on the other hand that the successful translations of neural MT systems sometimes bear resemblance with the translations of a rule-based MT system.
Original languageEnglish
Pages (from-to)159-170
Number of pages12
JournalPrague Bulletin of Mathematical Linguistics
Issue number1
Early online date6 Jun 2017
Publication statusPublished - Jun 2017


Dive into the research topics of 'A Linguistic Evaluation of Rule-Based, Phrase-Based, and Neural MT Engines'. Together they form a unique fingerprint.

Cite this