lr,4-1-P05-1074,bq task </term> . Previous work has used <term> monolingual parallel corpora </term> to extract and generate <term> paraphrases
other,11-1-P05-1074,bq corpora </term> to extract and generate <term> paraphrases </term> . We show that this task can be done
lr,9-2-P05-1074,bq show that this task can be done using <term> bilingual parallel corpora </term> , a much more commonly available <term>
lr,18-2-P05-1074,bq </term> , a much more commonly available <term> resource </term> . Using <term> alignment techniques
tech,1-3-P05-1074,bq available <term> resource </term> . Using <term> alignment techniques </term> from <term> phrase-based statistical
tech,4-3-P05-1074,bq <term> alignment techniques </term> from <term> phrase-based statistical machine translation </term> , we show how <term> paraphrases </term>
other,12-3-P05-1074,bq machine translation </term> , we show how <term> paraphrases </term> in one <term> language </term> can be
other,15-3-P05-1074,bq how <term> paraphrases </term> in one <term> language </term> can be identified using a <term> phrase
other,21-3-P05-1074,bq language </term> can be identified using a <term> phrase </term> in another language as a pivot .
other,3-4-P05-1074,bq language as a pivot . We define a <term> paraphrase probability </term> that allows <term> paraphrases </term>
other,7-4-P05-1074,bq paraphrase probability </term> that allows <term> paraphrases </term> extracted from a <term> bilingual parallel
lr,11-4-P05-1074,bq paraphrases </term> extracted from a <term> bilingual parallel corpus </term> to be ranked using <term> translation
other,18-4-P05-1074,bq parallel corpus </term> to be ranked using <term> translation probabilities </term> , and show how it can be refined
other,30-4-P05-1074,bq show how it can be refined to take <term> contextual information </term> into account . We evaluate our <term>
tech,3-5-P05-1074,bq </term> into account . We evaluate our <term> paraphrase extractio and ranking methods </term> using a set of <term> manual word alignments
other,12-5-P05-1074,bq ranking methods </term> using a set of <term> manual word alignments </term> , and contrast the <term> quality </term>
measure(ment),19-5-P05-1074,bq alignments </term> , and contrast the <term> quality </term> with <term> paraphrases </term> extracted
other,21-5-P05-1074,bq contrast the <term> quality </term> with <term> paraphrases </term> extracted from <term> automatic alignments
other,24-5-P05-1074,bq <term> paraphrases </term> extracted from <term> automatic alignments </term> . We present a <term> Czech-English
skrij podrobnosti