TAILIEUCHUNG - Báo cáo khoa học: "Machine Translation System Combination using ITG-based Alignments∗"

Given several systems’ automatic translations of the same sentence, we show how to combine them into a confusion network, whose various paths represent composite translations that could be considered in a subsequent rescoring step. We build our confusion networks using the method of Rosti et al. (2007), but, instead of forming alignments using the tercom script (Snover et al., 2006), we create alignments that minimize invWER (Leusch et al., 2003), a form of edit distance that permits properly nested block movements of substrings. . | Machine Translation System Combination using ITG-based Alignments Damianos Karakos Jason Eisner Sanjeev Khudanpur Markus Dreyer Center for Language and Speech Processing Johns Hopkins University Baltimore MD 21218 damianos eisner khudanpur dreyer @ Abstract Given several systems automatic translations of the same sentence we show how to combine them into a confusion network whose various paths represent composite translations that could be considered in a subsequent rescoring step. We build our confusion networks using the method of Rosti et al. 2007 but instead of forming alignments using the tercom script Snover et al. 2006 we create alignments that minimize invWER Leusch et al. 2003 a form of edit distance that permits properly nested block movements of substrings. Oracle experiments with Chinese newswire and weblog translations show that our confusion networks contain paths which are significantly better in terms of BLEU and TER than those in tercom-based confusion networks. 1 Introduction Large improvements in machine translation MT may result from combining different approaches to MT with mutually complementary strengths. System-level combination of translation outputs is a promising path towards such improvements. Yet there are some significant hurdles in this path. One must somehow align the multiple outputs to identify where different hypotheses reinforce each other and where they offer alternatives. One must then This work was partially supported by the DARPA GALE program Contract No HR0011-06-2-0001 . Also we would like to thank the IBM Rosetta team for the availability of several MT system outputs. use this alignment to hypothesize a set of new composite translations and select the best composite hypothesis from this set. The alignment step is difficult because different MT approaches usually reorder the translated words differently. Training the selection step is difficult because identifying the best hypothesis relative to a known reference .

• Báo cáo khoa học
• Machine Translation System Combination
• ITG based Alignments∗
• Damianos Karakos
• báo cáo khoa học
• báo cáo ngôn ngữ
• xử lý ngôn ngữ tự nhiên
• N gram based Statistical
• Machine Translation versus Syntax Augmented Machine Translation
• comparison and system combination
• scientific reports
• model language
• process natural language
• Boosting based System Combination
• Machine Translation
• Tong Xiao
• Jingbo Zhu
• Muhua Zhu
• ngôn ngữ tự nhiên
• Confusion Forest
• Taro Watanabe
• Comparative Study
• Hypothesis Alignment and its Improvement
• Improved Word Level
• System Combination
• mô hình ngôn ngữ
• Bài giảng Xử lý ngôn ngữ tự nhiên
• Kỹ thuật lập trình
• Xử lý nhập nhằng ngữ nghĩa
• Xử lý ngữ nghĩa
• Xử lý ngôn ngữ
• Natural Language Processing
• Ngôn ngữ lập trình
• Nhập nhằng cấu trúc
• Tri thức về ngôn ngữ
• Mô hình n gram
• Dịch máy
• Phương pháp dịch máy
• Hiểu ngôn ngữ
• Phân tích ngữ nghĩa
• Phân loại tin tự động
• Biểu diễn vị từ
• Thuộc tính về sự kiện
• Thuộc tính về sự kiện
• Mô hình ngôn ngữ Google Book N grams
• Mô hình ngôn ngữ KenLM
• Phương pháp làm mịn
• Tách từ tiếng Việt
• Gán nhãn từ loại
• Phân tích cú pháp
• Nghĩa từ vựng
• Phân giải nhập nhằng từ
• Phân loại văn bản
• Lexical chain
• Phản hồi thông tin
• Bài toán PTCP
• Cấu trúc ngữ pháp
• Phân tích cú pháp xác suất
• CKY kết hợp xác suất
• Văn phạm phi ngữ cảnh xác suất
• Hình thái học