It has been claimed that human translators rely on some sort of literal translation equivalences to produce translations and to check their validity. More effort would be required if translations are less literal. However, to our knowledge, there is no established metric to measure and quantify this claim. This paper attempts to bridge this gap by introducing a metric for measuring literality o...

We show that unseen words account for a large part of the translation error when moving to new domains. Using an extension of a recent approach to mining translations from comparable corpora (Haghighi et al., 2008), we are able to find translations for otherwise OOV terms. We show several approaches to integrating such translations into a phrasebased translation system, yielding consistent impr...

We present a systematic study of the effect of crowdsourced translations on Machine Translation performance. We compare Machine Translation systems trained on the same data but with translations obtained using Amazon’s Mechanical Turk vs. professional translations, and show that the same performance is obtained from Mechanical Turk translations at 1/5th the cost. We also show that adding a Mech...

In this paper, we propose a method for translating a given verb-noun collocation based on a parallel corpus and an additional monolingual corpus. Our approach involves two models to generate collocation translations. The combination translation model generates combined translations of the collocate and the base word, and filters translations by a target language model from a monolingual corpus,...

This paper describes a novel approach to syntactically-informed evaluation of machine translation (MT). Using a statistical, treebanktrained parser, we extract word-word dependencies from reference translations and then compile these dependencies into a representation that allows candidate translations to be evaluated by string comparisons, as is done in n-gram approaches to MT evaluation. This...

The present study aims at discussing whether metaphors in the Qur’an, revealed more than 1400 years ago, are dead, moribund or live and how these three types of metaphors have been translated in three English and three Persian translations of the Qur’an. The results reveal that among 70 metaphors examined, while only about 32.85% are live metaphors, about 67.14% are moribund, but none of the ca...

The purpose of this study was to compare and contrast postural responses to lateral and A/P surface translations by quantifying joint positions, bilateral three-dimensional (3-D) ground reaction forces, and lower limb and trunk muscle electromyographic (EMG) activity. Subjects stood on a movable platform which was randomly translated in four different directions. The kinematic patterns in respo...