基于混合语义空间的汉译英自动评分模型
|
摘要
针对传统翻译自动评分方法分析不够全面且准确率不高的问题,提出一种基于混合语义空间的汉译英自动评分模型。该模型通过规则与统计相结合的语法检错算法对待测译文进行语法分析,并基于混合语义空间的语义相关度算法对待测译文与标准译文进行分析,然后赋予相应的权重,从而对待测译文进行自动评分。实验结果表明,该方法与人工评分的平均误差仅为1.13,皮尔逊相关系数为0.87,具有较高的准确率。
Aiming at the case that traditional automatic correction method for English translation is not overall and in low accuracy,an automatic correction model of Chinese-English translation which is based on hybrid semantic space is studied and designed.The model can analyze the grammar errors of the Chinese-English essay by a syntactic error detection algorithm which combines rules and statistics,it also can compare the translation essay with the standard by a semantic relevancy algorithm which is based on hybrid semantic space.Meanwhile,the corresponding weights will be calculated.Then the efficient automatic scoring of the translation essay can get.Via experimenting with real data sets,the results show that the the average error between the proposed method and manual scoring is only 1.13,the pearson correlation coefficient of the method is 0.87 and the method has high accuracy and good practicability in automatic translation scoring field.
Aiming at the case that traditional automatic correction method for English translation is not overall and in low accuracy,an automatic correction model of Chinese-English translation which is based on hybrid semantic space is studied and designed.The model can analyze the grammar errors of the Chinese-English essay by a syntactic error detection algorithm which combines rules and statistics,it also can compare the translation essay with the standard by a semantic relevancy algorithm which is based on hybrid semantic space.Meanwhile,the corresponding weights will be calculated.Then the efficient automatic scoring of the translation essay can get.Via experimenting with real data sets,the results show that the the average error between the proposed method and manual scoring is only 1.13,the pearson correlation coefficient of the method is 0.87 and the method has high accuracy and good practicability in automatic translation scoring field.
引文
[1]周佳.浅谈2013新改革后大学英语六级考试的翻译技巧[J].海外英语,2014,18:63-64.
[2] DIKIS.An overview of automated scoring of essays[J].The Journal of Technology,Learing,and Assessment,2006,5(1):1-36.
[3] WAARD J D,NIDA E A.From One Language to Another:Functional Equivalence in Bible Translation[M].Nashville:Thomas Nelson Inc,1986:106-111.
[4] PAPINENI K,ROUKOS S,WARD T,et al.Bleu:a method for automatic evaluation of machine translation[R].Thomas J.Watson Research Center,2001:1-24.
[5]姚建民,周明,赵铁军,等.基于句子相似度的机器翻译评价方法及其有效性分析[J].计算机研究与发展,2004,41(7):1258-1265.
[6]文秋芳,秦颖,江进林.英语考试翻译自动评分中双语对齐技术的应用[J].外语电化教学,2009,125:3-8.
[7]田艳.网上英汉自动评分实践探索[J].中国翻译,2011,28(2):38-41.
[8] LI Li,WU W.Automated Chinese-English translation scoring based on answer knowledge base[C]//Proceedings of 12th IEEE International Conference on Cognitive Informatics&Cognitive Computing,2013:86-93.
[9]张英萍.大学英语四级新题型“段落翻译”探析[J].考试周刊,2014,(7):7-8.
[10] NABER D.A Rule-Based Style and Grammar Checker[M].Bielefeld:GRIN Verlag,2003.
[11] KNESER R,NEY H.Improved backing-off for M-gram language modeling[C]//IEEE International Conference on Acoustics,Speech,and Signal Processing,1995:181-184.
[12] MIKOLOV T,CHEN K,CORRADO G,et al.Efficient estimation of word representations in vector space[C]//Proceedings of Workshop at ICLR,2013:3111-3119.
[13] PENNINGTON J,SOCHER R,CHRISTOPHER D.Manning.GloVe:global vectors for word representation[C]//Proceedings of the 2014Conference on Empirical Methods in Natural Language Processing,2014:1532-1543.
[14] LIUH, SINGHP. ConceptNet:a practical commonsense reasoning toll-kit[J].BT Technology Journal,2004,22(4):211-226.
[15] GANITKEVITCH J,DURME B.V,CALLISONBURCH C.PPDB:the paraphrase database[C]//Proceedings of the 2013Conference of the North American Chapter of the Association for Computational Linguistics,2013:758-764.
[16] FARUQUI M,DODGE J,JAUHAR S.K,et al.Retrofitting word vectors to semantic lexicons[C]//Proceedings of NAACL,2015:1-10.
[17] SPEER R,CHIN J,HAVASI C.ConceptNet 5.5:an open multilingual graph of general Knowledge[J].AAAI,2017,(31):4444-4451.
[18] ARORA S,LIANG Y Y,MA T Y.A simple but toughto-beat baseline for sentence embeddings[C]//ICLR Conference,2017:1-16.
[2] DIKIS.An overview of automated scoring of essays[J].The Journal of Technology,Learing,and Assessment,2006,5(1):1-36.
[3] WAARD J D,NIDA E A.From One Language to Another:Functional Equivalence in Bible Translation[M].Nashville:Thomas Nelson Inc,1986:106-111.
[4] PAPINENI K,ROUKOS S,WARD T,et al.Bleu:a method for automatic evaluation of machine translation[R].Thomas J.Watson Research Center,2001:1-24.
[5]姚建民,周明,赵铁军,等.基于句子相似度的机器翻译评价方法及其有效性分析[J].计算机研究与发展,2004,41(7):1258-1265.
[6]文秋芳,秦颖,江进林.英语考试翻译自动评分中双语对齐技术的应用[J].外语电化教学,2009,125:3-8.
[7]田艳.网上英汉自动评分实践探索[J].中国翻译,2011,28(2):38-41.
[8] LI Li,WU W.Automated Chinese-English translation scoring based on answer knowledge base[C]//Proceedings of 12th IEEE International Conference on Cognitive Informatics&Cognitive Computing,2013:86-93.
[9]张英萍.大学英语四级新题型“段落翻译”探析[J].考试周刊,2014,(7):7-8.
[10] NABER D.A Rule-Based Style and Grammar Checker[M].Bielefeld:GRIN Verlag,2003.
[11] KNESER R,NEY H.Improved backing-off for M-gram language modeling[C]//IEEE International Conference on Acoustics,Speech,and Signal Processing,1995:181-184.
[12] MIKOLOV T,CHEN K,CORRADO G,et al.Efficient estimation of word representations in vector space[C]//Proceedings of Workshop at ICLR,2013:3111-3119.
[13] PENNINGTON J,SOCHER R,CHRISTOPHER D.Manning.GloVe:global vectors for word representation[C]//Proceedings of the 2014Conference on Empirical Methods in Natural Language Processing,2014:1532-1543.
[14] LIUH, SINGHP. ConceptNet:a practical commonsense reasoning toll-kit[J].BT Technology Journal,2004,22(4):211-226.
[15] GANITKEVITCH J,DURME B.V,CALLISONBURCH C.PPDB:the paraphrase database[C]//Proceedings of the 2013Conference of the North American Chapter of the Association for Computational Linguistics,2013:758-764.
[16] FARUQUI M,DODGE J,JAUHAR S.K,et al.Retrofitting word vectors to semantic lexicons[C]//Proceedings of NAACL,2015:1-10.
[17] SPEER R,CHIN J,HAVASI C.ConceptNet 5.5:an open multilingual graph of general Knowledge[J].AAAI,2017,(31):4444-4451.
[18] ARORA S,LIANG Y Y,MA T Y.A simple but toughto-beat baseline for sentence embeddings[C]//ICLR Conference,2017:1-16.