Đang chuẩn bị nút TẢI XUỐNG, xin hãy chờ
Tải xuống
Trong hệ thống diễn giải hiện nay có các tùy chọn dịch votar và điện tử và 'en votar mà các bộ giải mã trước đây đã không có lựa chọn. e 7.1.3.2 hành vi trên các từ và cụm từ trước đây chưa từng | 128 Chapter 7. Translation Experiments votaré en favor de la aprobación vote voting 5 favour 5 I i shall vote ivotea to vote in i agree with in on at into favour in favour of - .for- from I favour of favour of the ------ior ---------- Tn favoUr for Of the the from the in the the approval to the f diScharge S J passing of the adoption theapproval the discharge the passing Figure 7.3 In the paraphrase system there are now translation options for votare and and votare en for which the decoder previously had no options. 7.1.3.2 Behavior on previously unseen words and phrases The expanded phrase table of the paraphrase system results in different behavior for unknown words and phrases. Now the decoder has access to a wider range of translation options as illustrated in Figure 7.3. For unknown words and phrases for which no paraphrases were found or whose paraphrases did not occur in the baseline phrase table the behavior of the paraphrase system is identical to the baseline system. We did not generate paraphrases for names numbers and foreign language words since these items should not be translated. We manually created a list of the nontranslating words from the test set and excluded them from being paraphrased. 7.1.3.3 Additional feature function In addition to expanding the phrase table we also augmented the paraphrase system by incorporating the paraphrase probability into an additional feature function that was not present in the baseline system as described in Section 5.4.2. We calculated paraphrase probabilities using the definition given in Equation 3.6. This definition allowed us to assign improved paraphrase probabilities by calculating the probability using multiple parallel corpora. We omitted other improvements to the paraphrase probability described in Chapter 4 including word sense disambiguation and re-ranking paraphrases based on a language model probability. These were omitted simply as a matter of convenience and their inclusion might have resulted in
