Analysing the Impact of Interactive Machine Translation on Post-editing Effort

A user study of neural interactive translation prediction

“…Koehn (2009) found that, relative to unassisted translation, both ITP and PE produced better quality, faster translations, but that ITP did not yet yield time gains to the level of PE. Similar findings of translators being slower overall in ITP than in PE are reported in Underwood et al (2014), Green et al (2014), Sanchis-Trilles et al (2014), Alabau et al (2016) and Alves et al (2016). The number of participants in these research studies ranged from 5 to 32; language pairs investigated were English to Spanish, Portuguese, and German, and French to English.…”

“…Both Macklovitch (2006) and Koehn (2009) found ITP to be an efficient alternative to unassisted translation in terms of processing time. So far, phrase-based statistical ITP has not yet proven to be faster than PE (Koehn 2009;Sanchis-Trilles et al 2014;Underwood et al 2014;Green et al 2014;Alves et al 2016;Alabau et al 2016). In this paper we present the results of an empirical study on translation productivity in ITP with an underlying neural MT system (NITP).…”

Assessing complexity and difficulty levels of machine-translated texts

“…Since comparability may be an issue in studies comparing performance in the translation of two or more texts, the predominance of studies involving only one text, as pointed out by Krings (2001), could be a potential explanation for the little attention heeded to text selection by that time.1 9 0 F 1 However, a lack of grounded description of text selection holds true in present days, even in studies involving two or more texts (ALVES et al, 2016, MOORKENS et al, 2015. Consequently, after reading a TPR study, readers usually wonder how the researchers chose the text(s) to carry out their experiments and how they assessed the level of difficulty or complexity for translation.1 9 1 F 2 O'Brien (2013, p. 117) also notes this lack of adequate operationalization of text selection procedures and claims the use of readability indexes to measure text translatability is "still highly questionable".…”

“…Results of the experiment itself, which included four different monolingual post-editing tasks, are available in Fonseca (2016). 2) How can we compare word frequency to word familiarity as proposed by Jensen (2009)? 3) How can we assess text difficulty and complexity without assessing the texts ourselves?…”

Cognitive effort and different task foci in post-editing of machine translation: A think-aloud study

“…As in recent post-editing research (e.g. Alves et al 2016), fixation count and the average duration of eye fixations were the specific eye-tracking metrics used here. The use of subjective ratings to estimate effort, in turn, is based on the assumption that individuals are able to report on cognitive (or mental) effort expenditure in terms of numerical scores (see O'Donnel and Eggemeier 1986;Paas 1992).…”