Explicit length modelling for statistical machine translation

作     者：Albert Silvestre-Cerda, Joan Andres-Ferrer, Jesus Civera, Jorge 

作者机构：Univ Politecn Valencia Dept Sistemes Informat & Computacio Valencia 46022 Spain 

出 版 物：《PATTERN RECOGNITION》 (图形识别)

年 卷 期：2012年第45卷第9期

页      面：3183-3192页

核心收录：

学科分类：0808[工学-电气工程] 08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：EC (FEDER/FSE) under the transLectures project [FP7-ICT-2011-7-287755] Spanish MEC/MICINN [CSD2007-00018, TIN2009-14511] FPU [AP2010-4349] Spanish MITyC [TSI-020110-2009-439] Generalitat Valenciana [Prometeo/2009/014, GV/2010/067] UPV 

主　　题：Length modelling Log-linear models Phrase-based models Statistical machine translation 

摘      要：Explicit length modelling has been previously explored in statistical pattern recognition with successful results. In this paper, two length models along with two parameter estimation methods and two alternative parametrisations for statistical machine translation (SMT) are presented. More precisely, we incorporate explicit bilingual length modelling in a state-of-the-art log-linear SMT system as an additional feature function in order to prove the contribution of length information. Finally, a systematic evaluation on reference SMT tasks considering different language pairs proves the benefits of explicit length modelling. (C) 2012 Elsevier Ltd. All rights reserved.