高考语文阅读理解问答相对普通阅读理解问答难度更大,问句抽象表述的理解需要更深层的语言分析技术,答案候选句抽取更注重与问句的关联分析,答案候选句排序更注重答案句之间的语义相关性。为此,该文提出借助框架语义匹配和框架语义关系抽取答案候选句,在排序时引入流形排序模型,通过答案句之间的框架语义相关度将排序分数进行传播,最终选取分数较高的Top-4作为答案句。在北京近12年高考语文阅读理解问答题上的准确率与召回率分别达到了53.65%与79.06%。
Abstract
Reading comprehension QA for Chinese College Entrance Examination is much more difficult than general reading comprehension QA in that it requires deeper linguistic analysis technology to understand the question, and the semantic correlation between the answers and questions. This paper proposes to extract the candidate answer sentences by frame semantic match and frame-frame semantic relation, and the manifold-ranking model are applied to propogate the frame semantic relevancy to decide the top-four candidate answers. The accuracy and recall on the college entrance examination of Beijing in recent twelve years is 53.65% and 79.06%, respectively.
关键词
高考语文 /
阅读理解 /
框架语义 /
答案句抽取 /
流形排序
{{custom_keyword}} /
Key words
College Entrance Examination on Chinese /
reading comprehension /
frame semantic /
answers sentences extraction /
Manifold-ranking
/
/
/
/
/
/
/
/
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 黄昌宁.IBM深度问答系统战胜顶尖人类选手所想到的[J].中文信息学报,2011, 25(6): 21-25.
[2] 余正涛,樊孝忠,郭剑毅,耿增民.基于潜在语义分析的汉语问答系统答案提取[J]. 计算机学报, 2006, 29(10): 1889-1893.
[3] 吴友政,赵军,徐波.基于主题语言模型的句子检索算法[J]. 计算机研究与发展, 2007, 44(2): 288-295.
[4] 刘群,李素建.基于《知网》的词汇语义相似度[J].中文计算语言学,2002, 7(2): 59-76.
[5] Li Ru, Li Shuanghong, Zhang Zezheng. The SemanticComputing Model of Sentence Similarity Based on Chinese FrameNet[C]//Proceedings of the 2009 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology. IEEE Computer Society, 2009: 255-258.
[6] 李茹,王智强,李双红,等.基于框架语义分析的汉语句子相似度计算[J].计算机研究与发展, 2013, 50(8): 1728-1736.
[7] 田久乐,赵蔚.基于同义词词林的词语相似度计算方法[J].吉林大学学报: 信息科学版,2010,28(6): 602-608.
[8] 张志昌, 张宇, 刘挺,等. 基于浅层语义树核的阅读理解答案句抽取[J].中文信息学报, 2008, 22(1): 80-86.
[9] Higashinaka R, Isozaki H. Corpus-based Question Answering for Why-questions[C]//Proceedings of IJCNLP. Hyderabad, India: AFNLP, 2008: 418-425.
[10] 张志昌,张宇,刘挺,等.基于话题和修辞识别的阅读理解why型问题回答[J].计算机研究与发展, 2011, 54(2): 216-223.
[11] 王智强,李茹,梁吉业,等.基于汉语篇章框架语义分析的阅读理解问答研究[J].计算机学报, 2016,39(04): 795-807.
[12] Wang H, Bansal M, Gimpel K, et al. Machine ComprehensionwithSyntax, Frames,andSemantics[C]//Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 2: Short Papers). 2015.
[13] Riloff E, Thelen M. Rule-based Question Answering System for Reading Comprehension Tests[J]. Workshop on Reading Comprehension, NAACL/ANLP-2000, 2000.
[14] Hao X, Chang X, Liu K. A Rule-based Chinese Question Answering System for Reading Comprehension Tests[C]//Proceedings of the International Conference on Intelligent Information Hiding and Multimedia Signal Processing. IEEE, 2007: 325-329.
[15] Akour M, Abufardeh S, Magel K, et al. QArabPro: ARule Based Question Answering System for Reading Comprehension Tests in Arabic[J]. American Journal of Applied Sciences, 2011, 8(6): 652-661.
[16] 由丽萍. 构建现代汉语框架语义知识库技术研究[D]. 上海师范大学, 2006.
[17] 郝晓燕, 刘伟, 李茹,等. 汉语框架语义知识库及软件描述体系[J]. 中文信息学报, 2007, 21(5): 96-100.
[18] 郝晓燕. 汉语框架网络工程构建及应用[M]. 电子工业出版社, 2011.
[19] Zhou, Dengyong, Weston, et al. Ranking on DataManifolds[J]. Advances in Neural InformationProcessing Systems, 2003: 169-176.
[20] 王荣波, 池哲儒, 常宝宝,等. 基于词串粒度及权值的汉语句子相似度衡量[J]. 计算机工程, 2005, 31(13): 142-144.
[21] 刘敏. 基于词向量的句子相似度计算及其在基于实例的机器翻译中的应用[D]. 北京理工大学, 2015.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家863计划(2015AA015407);国家自然科学基金(61373082,61502287,61673248);山西省科技基础条件平台建设项目(2014091004-0103);山西省回国留学人员科研资助项目(2013-015);中国民航大学信息安全测评中心开放课题基金(CAAC-ISECCA-201402);山西省高校科技创新项目(201505)
{{custom_fund}}
PDF(4275 KB)
