Efficient beam search decoder for phrase-based statistical machine translation

• Artificial intelligence • Previous Articles Next Articles

Efficient beam search decoder for phrase-based statistical machine translation

Received:2007-02-15 Revised:1900-01-01 Online:2007-08-01 Published:2007-08-01
Contact: Miao Li

一种基于短语统计机器翻译的高效柱搜索解码器

罗毅李淼张建

中国科学院合肥智能机械研究所;中国科学技术大学信息科学技术学院中国科学院合肥智能机械研究所;中国科学技术大学信息科学技术学院中国科学院合肥智能机械研究所;中国科学技术大学信息科学技术学院

通讯作者: 李淼

Abstract

Abstract: An efficient beam search decoder for phrase-based statistical machine translation was described. The efficiency of search algorithm is the key to decoding process. After introducing the conventional beam search decoding algorithm, some efficiency improving measures were proposed. Dynamic pruning strategy enhanced the accuracy of pruning by improving that the original fixed pruning had not enough response to the current situation of search. Pre-pruning strategy was used to limit the poor sprawl, reduce unnecessary expansion and improve search speed. A rapid reordering constrains strategy was presented based on the research of the current major reordering constrains. In addition, the domain term always has the only translation, so a special process approach was put forward to improve the accuracy of the translation. Comparative analysis of the experimental results proves the effectiveness of the algorithm.

Key words: statistical machine translation, decoder, beam search, dynamic pruning, reordering constrains

摘要： 描述了一种基于短语统计机器翻译的柱搜索解码器。搜索算法的效率是解码的关键，基于传统的柱搜索解码算法，提出了提高搜索效率的改进措施：动态剪枝策略改进了原来固定地剪枝对搜索当前情形反应不足的问题，提高了剪枝精度；预剪枝策略限制了较差的扩展，减少了不必要的扩展，提高了搜索速度；在研究了当前主要位置重排限制的基础上，提出了一种快速位置重排限制策略，加快了位置重排时的解码速度。此外，针对领域术语翻译唯一性问题提出了专门处理方法以提高翻译的准确度。分析对比实验结果，证明了算法的有效性。

关键词: 统计机器翻译, 解码器, 柱搜索, 动态剪枝, 位置重排限制

Yi Luo Miao Li Jian Zhang. Efficient beam search decoder for phrase-based statistical machine translation[J]. Journal of Computer Applications.

罗毅李淼张建. 一种基于短语统计机器翻译的高效柱搜索解码器[J]. 计算机应用.

[1]	ZHANG Mingming, LU Qingning, LI Wenzhong, SONG Hu. Deep neural network compression algorithm based on combined dynamic pruning [J]. Journal of Computer Applications, 2021, 41(6): 1589-1596.
[2]	LIU Shize, QIN Yanjun, WANG Chenxing, SU Lin, KE Qixue, LUO Haiyong, SUN Yi, WANG Baohui. Traffic flow prediction algorithm based on deep residual long short-term memory network [J]. Journal of Computer Applications, 2021, 41(6): 1566-1572.
[3]	DU Xixi, CHENG Hua, FANG Yiquan. Reinforced automatic summarization model based on advantage actor-critic algorithm [J]. Journal of Computer Applications, 2021, 41(3): 699-705.
[4]	Yang ZHAO, Bo LI. Single image dehazing based on conditional generative adversarial network with enhanced generator [J]. Journal of Computer Applications, 2021, 41(12): 3686-3691.
[5]	DING Xiangguo, SANG Jitao. Joint extraction of entities and relations based on relation-adaptive decoding [J]. Journal of Computer Applications, 2021, 41(1): 29-35.
[6]	CHEN Xiukai, LU Zhihua, ZHOU Yu. Speech separation algorithm based on convolutional encoder decoder and gated recurrent unit [J]. Journal of Computer Applications, 2020, 40(7): 2137-2141.
[7]	YANG Wenxia, WANG Meng, ZHANG Liang. Semantic face image inpainting based on U-Net with dense blocks [J]. Journal of Computer Applications, 2020, 40(12): 3651-3657.
[8]	JIA Ruiming, QIU Zhenzhi, CUI Jiali, WANG Yiding. Deep multi-scale encoder-decoder convolutional network for blind deblurring [J]. Journal of Computer Applications, 2019, 39(9): 2552-2557.
[9]	ZHANG Xiang-xian YANG Tao WEI Dong-mei XIANG Ling. Design and FPGA implementation of parallel high-efficiency BCH decoder [J]. Journal of Computer Applications, 2012, 32(03): 867-869.
[10]	. High-throughput Reed-Solomon decoder [J]. Journal of Computer Applications, 2008, 28(10): 2696-2700.
[11]	. Design and implementation of JPEG image decoder in embedded systems [J]. Journal of Computer Applications, 2007, 27(9): 2324-2326.
[12]	;;;. Removing the feature of instruction bundle in IA-64 binary translation [J]. Journal of Computer Applications, 2006, 26(5): 1186-1189.
[13]	XING Lu-lu,XIE Sha-sha,SUN Jia-song,WANG Zuo-ying. Application of on-line filler modeling in specific domain speech recognition system [J]. Journal of Computer Applications, 2005, 25(12): 2795-2797.