What is an error and what is a translation ‘error’? The term “error” usually means that something is wrong. In written texts – both in original texts and translated texts – errors can be classified as, for example, pragmatic, semantic, idiomatic, orthographic, linguistic or stylistic errors. But what is a ‘translation error’ – with focus on ‘translation’? If we define a translation as the produ...

In this talk I am concerned with the following questions: 1. What is phonological typology? 2. How are phonological typology and phonetic typology the same/different? 3. How are phonological typology and general phonology the same/different? 4. How are phonological typology and general typology the same/different? Despite earlier work by Trubetzkoy, Jakobson, Martinet, Greenberg and others, and...

We investigate the use of error classification results for automatic evaluation of machine translation output. Five basic error classes are taken into account: morphological errors, syntactic (reordering) errors, missing words, extra words and lexical errors. In addition, linear combinations of these categories are investigated. Correlations between the class error rates and human judgments are...

Machine translation (MT) systems are not able to always produce translations of human-level quality. As a practical means of such MT systems, we investigated the potential of pre-editing strategy, by collecting actual pre-edit instances using a human-in-the-loop protocol. In our study, targeting Japanese-to-English translation on four different datasets and using an offthe-shelf MT system, we c...

This talk discusses some o f the challenges posed for theoretical linguistics by transla­ tional problems. It briefly discusses different aspects o f the concept o f translational equivalence and the degree to which this relation may be captured by machine translation (MT) systems. It is argued that the formal requirements and multilingual perspective of MT provide new criteria o f adequacy for...

Among the applications of language technology to human translation, little attention has been paid to error detection. This article discusses the issue from three perspectives. In general terms, it considers the nature of translation errors and what might be involved in detecting them. On a more specific level, it identifies a number of error types which meet the twin criteria of feasibility (e...

Evaluation and error analysis of machine translation output are important but difficult tasks. In this work, we propose a novel method for obtaining more details about actual translation errors in the generated output by introducing the decomposition of Word Error Rate (WER) and Position independent word Error Rate (PER) over different Partof-Speech (POS) classes. Furthermore, we investigate tw...

the present study assessed the construct validity and reliability of a researcher-constructed psycho-motor mechanism scale based on the communicative theory of translation proposed by pacte (2003). in doing so, the necessary criteria for designing the scale were obtained by a thorough review of related literature on previously constructed scales in error analysis or holistic ones. moreover, in ...

This paper presents a translation model that is based on tree sequence alignment, where a tree sequence refers to a single sequence of subtrees that covers a phrase. The model leverages on the strengths of both phrase-based and linguistically syntax-based method. It automatically learns aligned tree sequence pairs with mapping probabilities from word-aligned biparsed parallel texts. Compared wi...

There are above 60 million first language speakers of Urdu and above 104 million second language speakers. Lot of knowledge on the internet available/useful to these speakers of Urdu is in English. The contrast in typology of both languages is interesting to study for Statistical Machine Translation. However, there is almost no parallel aligned data available freely for the selected language pa...