| measure(ment),2-1-E06-1031,ak | context </term> . Most state-of-the-art <term> | evaluation measures | </term> for <term> machine translation </term> | #11262 Most state-of-the-artevaluation measures for machine translation assign high costs to movements of word blocks. | |
| tech,5-1-E06-1031,ak | <term> evaluation measures </term> for <term> | machine translation | </term> assign high <term> costs </term> to <term> | #11265 Most state-of-the-art evaluation measures formachine translation assign high costs to movements of word blocks. | |
| measure(ment),9-1-E06-1031,ak | machine translation </term> assign high <term> | costs | </term> to <term> movements </term> of <term> word | #11269 Most state-of-the-art evaluation measures for machine translation assign highcosts to movements of word blocks. | |
| other,11-1-E06-1031,ak | </term> assign high <term> costs </term> to <term> | movements | </term> of <term> word blocks </term> . In many | #11271 Most state-of-the-art evaluation measures for machine translation assign high costs tomovements of word blocks. | |
| other,13-1-E06-1031,ak | costs </term> to <term> movements </term> of <term> | word blocks | </term> . In many cases though such <term> | #11273 Most state-of-the-art evaluation measures for machine translation assign high costs to movements ofword blocks. | |
| other,5-2-E06-1031,ak | </term> . In many cases though such <term> | movements | </term> still result in correct or almost | #11281 In many cases though suchmovements still result in correct or almost correct sentences. | |
| other,13-2-E06-1031,ak | result in correct or almost correct <term> | sentences | </term> . In this paper , we will present | #11289 In many cases though such movements still result in correct or almost correctsentences. | |
| measure(ment),9-3-E06-1031,ak | this paper , we will present a new <term> | evaluation measure | </term> which explicitly models <term> block | #11300 In this paper, we will present a newevaluation measure which explicitly models block reordering as an edit operation. | |
| other,14-3-E06-1031,ak | measure </term> which explicitly models <term> | block reordering | </term> as an <term> edit operation </term> . | #11305 In this paper, we will present a new evaluation measure which explicitly modelsblock reordering as an edit operation. | |
| other,18-3-E06-1031,ak | <term> block reordering </term> as an <term> | edit operation | </term> . Our <term> measure </term> can be exactly | #11309 In this paper, we will present a new evaluation measure which explicitly models block reordering as anedit operation. | |
| measure(ment),1-4-E06-1031,ak | an <term> edit operation </term> . Our <term> | measure | </term> can be exactly calculated in <term> | #11313 Ourmeasure can be exactly calculated in quadratic time. | |
| other,7-4-E06-1031,ak | </term> can be exactly calculated in <term> | quadratic time | </term> . Furthermore , we will show how | #11319 Our measure can be exactly calculated inquadratic time. | |
| measure(ment),7-5-E06-1031,ak | Furthermore , we will show how some <term> | evaluation measures | </term> can be improved by the introduction | #11329 Furthermore, we will show how someevaluation measures can be improved by the introduction of word-dependent substitution costs. | |
| measure(ment),5-6-E06-1031,ak | </term> . The correlation of the new <term> | measure | </term> with <term> human judgment </term> has | #11347 The correlation of the newmeasure with human judgment has been investigated systematically on two different language pairs. | |
| other,7-6-E06-1031,ak | of the new <term> measure </term> with <term> | human judgment | </term> has been investigated systematically | #11349 The correlation of the new measure withhuman judgment has been investigated systematically on two different language pairs. | |
| other,16-6-E06-1031,ak | investigated systematically on two different <term> | language pairs | </term> . The experimental results will show | #11358 The correlation of the new measure with human judgment has been investigated systematically on two differentlanguage pairs. | |
| measure(ment),4-8-E06-1031,ak | correlation . Results from experiments with <term> | word dependent substitution costs | </term> will demonstrate an additional increase | #11380 Results from experiments withword dependent substitution costs will demonstrate an additional increase of correlation between automatic evaluation measures and human judgment. | |
| other,20-8-E06-1031,ak | automatic evaluation measures </term> and <term> | human judgment | </term> . In this paper , we investigate | #11396 Results from experiments with word dependent substitution costs will demonstrate an additional increase of correlation between automatic evaluation measures andhuman judgment. |
