结合排序学习的趣味成语生成模型
|
摘要
汉语中的成语语言精练,典故源远流长,其趣味性是成语研究的一个重要方面.本文依据语音模板生成候选成语集合,从中提取语音、幽默、语义、情感和形态五大类11个特征,将这些特征融入到排序学习的相关算法中,从候选成语集合中检索趣味成语,进而构建趣味成语的生成模型.该模型将成语生成问题映射到信息检索领域,以查询及相关反馈的技术解决生成问题.经机器和人工的双重评估,实验结果表明五个维度的特征能够细致刻画趣味成语,生成质量较高,模型具有一定的实用价值.
Chinese idioms feature a concise form,most of which come from centuries-old allusions. Interestingness of idioms is an important aspect of idiom research. In this paper,we use phonetic templates to generate candidate sets of idioms,and extract eleven features of these idioms respectively from five dimensions,namely,phonetics,humor,semantics,sentiment and morphology. We integrate these features into the relevant algorithms of learning to rank so as to retrieve interesting idioms from the foregoing candidate set for an idiom generation model. The learned model transforms the problem of idiom generation into an information retrieval problem,and uses query and relevance feedback technology to solve the generation problem. Based on both manual and automatic evaluation,experimental results showthat interesting idioms can be precisely described and screened out based on these five-dimensional features. The idiom-generating model proves to have high quality and certain practical value.
Chinese idioms feature a concise form,most of which come from centuries-old allusions. Interestingness of idioms is an important aspect of idiom research. In this paper,we use phonetic templates to generate candidate sets of idioms,and extract eleven features of these idioms respectively from five dimensions,namely,phonetics,humor,semantics,sentiment and morphology. We integrate these features into the relevant algorithms of learning to rank so as to retrieve interesting idioms from the foregoing candidate set for an idiom generation model. The learned model transforms the problem of idiom generation into an information retrieval problem,and uses query and relevance feedback technology to solve the generation problem. Based on both manual and automatic evaluation,experimental results showthat interesting idioms can be precisely described and screened out based on these five-dimensional features. The idiom-generating model proves to have high quality and certain practical value.
引文
[1]Huang Bo-rong,Liao Xu-dong.The modern Chinese[M].Beijing:Higher Education Press,2011.
[2]Xinhua idiom dictionary[M].Beijing:The Commercial Press,2002.
[3]Igor Labutov,Hod Lipson.Humor as circuits in semantic networks[C].Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics,Jeju Korea,2012:150-155.
[4]Alessandro Valitutt,Hannu Toivonen.Let everything turn well in your w ife:generation of adult humor using lexical constraints[C].Proceedings of the 51st Annual M eeting of the Association for Computational Linguistics,Sofia,Bulgaria,2013:243-248.
[5]Zhang Dong-yu,Yang Liang,Zheng Pu-qi,et al.Construction and application of affective metaphor corpus[J].Scientia Sinica(Informationis),2015,12(45):1574-1587.
[6]Lin Hong-fei,Zhang Dong-yu,Yang Liang,et al.Computational humor researches and application[J].Journal of Shandong University,2016,7(51):1-10.
[7]Dario Bertero,Pascale Fung.A long short-term memory framework for predicting humor in dialogues[C].Proceedings of North American Chapter of the Association for Computational Linguistics,San Diego,CA,2016:130-135.
[8]Zeng Xiao-bing,Zhang Zhi-ping,Liu Rong.Investigation and discussion on Chinese idioms and idiomatic phrases in the Chinese language situation report[J].Journal of Chinese Information Processing,2008,22(6):43-49.
[9]Xu Yao-min.Stereotypes,definitions and demarcation of idioms[J].Studies of the Chinese Language,1997,46(1):11-17.
[10]Ni Bao-yuan,Yao Peng-ci.Four questions of synonymous idiom[J].Studies of the Chinese Language,1995,44(1):23-28.
[11]Luo Ling,Chen Yi-dong,Shi Xiao-dong,et al.Chinese idiom translation based on paragraphing[J].Journal of Chinese Information Processing,2015,29(4):166-174.
[12]Ran Jie,Sun Yu,Chang Xia.Research of construction of the idiom story ontology based on OWL[J].Computer Technology and Development,2010,20(5):63-66.
[13]Yang Xue-song,Ran Jie.Research of semantic retrieval model based on idiom literary quotation ontology[J].Computer Science,2011,38(10A):219-221.
[14]Zhang Pu.Thinking about language sense and circulation[J].Language Teaching and Linguistic Studies,1999,21(2):83-96.
[15]Lv Shu-xiang.A preliminary study on the problem of single and double syllables in modern Chinese[J].Studies of the Chinese Language,1963,12(1):10-22.
[16]Li Xu.Humor and humor creation[J].Literary Review,1991,33(4):152-154.
[17]Fu Yun.The use of rhetoric in humor[J].Journal of Zhengzhou University,2005,38(4):155-156.
[18]Bu Zhan-ting.Study on the attitude of hyperbole[J].Contemporary Rhetoric,2010,29(4):53-59.
[19]Qiu Jin,Liao Qiao-yun.Hyperbole research under mind philosophy[J].Contemporary Rhetoric,2014,33(4):74-80.
[20]Zhu Yang.Specific case of humorous utterance translation from the perspective of relevance theory[J].Foreign Languages and Their Teaching,2007,29(2):57-60
[21]Xu Lin-hong,Lin Hong-fei,Pan Yu.Constructing the affective lexcion ontology[J].Journal of the China Society for Scientific and Technical Information,2008,27(2):180-185.
[22]Qian Kai-yu,Guo Li-peng.Research on sentiment analysis of social netw ork review s w ith idioms[J].Journal of Chinese Computer Systems,2017,6(38):1273-1277.
[23]Dictionary of the idiom[M].Beijing:The Commercial Press,2017.
[24]Zeng Jie-ying,Yu Bai-lin.The effects of Chinese characters'construction under thetachistoscopic repetition and non-tachistoscopic conditions[J].Psychological Science,1999,36(22):305-309.
[25]Gao Jia-ying.Modern Chinese calligraphy[M].Beijing:Higher Education Press,1993.
[26]Yu Bai-lin,Cao He-xi.A new exploration on the effect of strokenumber in identification of Chinese characters[J].Acta Psychologica Sinica,1992,37(2):120-126.
[27]Cao Chuan-yong,Shen Ye.The recognition of Chinese characters under tachistoscopic condition in primary school children[J].Acta Psychologica Sinica,1963,8(4):271-279.
[28]Hofmann K,Whiteson S,de Rijke M.Balancing exploration and exploitation in listwise and pairwise online learning to rank for information retrieval[J].Information Retrieval,2013,16(1):63-90.
[29]Wu Q,Burges C J C,Svore K M.Adapting boosting for information retrieval measures[J].Information Retrieval,2010,13(3):254-270.
[30]Liu Chang-zheng,Qin Peng.An investigation of idioms in Chinese new spapers based on the dynamic circulation corpus(DCC)[J].Applied Linguistics,2007,16(3):78-86.
[31]Zhang Tie-wen.The number and generation of idioms[J].Language Planning,1999,44(5):23-24.
[32]Xu Zhao-ben.Idiom knowledge introduction[M].Beijing:Bejing Press,1980.
[1]黄伯荣,廖序东.现代汉语[M].北京:高等教育出版社,2011.
[2]新华成语词典[M].北京:商务印书馆,2002.
[5]张冬瑜,杨亮,郑朴琪,等.情感隐喻语料库构建与应用[J].中国科学,2015,12(45):1574-1587.
[6]林鸿飞,张冬瑜,杨亮,等.幽默计算及其应用研究[J].山东大学学报,2016,7(51):1-10.
[8]曾小兵,张志平,刘荣.《中国语言生活状况报告》中成语与习语的调查与思考[J].中文信息学报,2008,22(6):43-49.
[9]徐耀民.成语的划界、定型和释义问题[J].中国语文,1997,46(1):11-17.
[10]倪宝元,姚鹏慈.等义成语四题[J].中国语文,1995,44(1):23-28.
[11]罗凌,陈毅东,史晓东,等.基于复述技术的汉语成语翻译方法研究[J].中文信息学报,2015,29(4):166-174.
[12]冉婕,孙瑜,昌霞.基于OWL的成语典故本体构建研究[J].计算机技术与发展,2010,20(5):63-66.
[13]杨雪松,冉婕.基于成语典故本体的语义检索模型研究[J].计算机科学,2011,38(10A):219-221.
[14]张普.关于语感与流通度的思考[J].语言教学与研究,1999,21(2):83-96.
[15]吕叔湘.现代汉语单双音节问题初探[J].中国语文,1963,12(1):10-22.
[16]李旭.幽默和幽默创造[J].文学评论,1991,33(4):152-154.
[17]付云.幽默中修辞手法的运用[J].郑州大学学报,2005,38(4):155-156.
[18]布占廷.夸张修辞的态度意义研究[J].当代修辞学,2010,29(4):53-59.
[19]邱晋,廖巧云.心智哲学视域下的夸张修辞研究[J].当代修辞学,2014,33(4):74-80.
[20]朱燕.关联理论对幽默言语翻译的诠释力[J].外语与外语教学,2007,29(2):57-60.
[21]徐琳宏,林鸿飞,潘宇.情感词汇本体的构造[J].情报学报,2008,27(2):180-185.
[22]钱凯雨,郭立鹏.融入习语信息的网络评论情感分析研究[J].小型微型计算机系统,2017,6(38):1273-1277.
[23]成语大词典[M].北京:商务印书馆,2017.
[24]曾捷英,喻柏林.速示重复和非速示呈现下的汉字结构方式效应[J].心理科学,1999,36(22):305-309.
[25]高家莺.现代汉字学[M].北京:高等教育出版社,1993.
[26]喻柏林,曹河析.汉字识别中的笔画数效应新探-兼论字频效应[J].心理学报,1992,37(2):120-126.
[27]曹传詠,沈晔.在速示条件下儿童辨认汉字字形的试探性研究[J].心理学报,1963,8(4):271-279.
[30]刘长征,秦鹏.基于中国主流报纸动态流通语料库(DCC)的成语使用情况调查[J].语言文字应用,2007,16(3):78-86.
[31]张铁文.成语的数量及产生年代[J].语文建设,1999,44(5):23-24.
[32]许肇本.成语知识浅谈[M].北京:北京出版社,1980.
[2]Xinhua idiom dictionary[M].Beijing:The Commercial Press,2002.
[3]Igor Labutov,Hod Lipson.Humor as circuits in semantic networks[C].Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics,Jeju Korea,2012:150-155.
[4]Alessandro Valitutt,Hannu Toivonen.Let everything turn well in your w ife:generation of adult humor using lexical constraints[C].Proceedings of the 51st Annual M eeting of the Association for Computational Linguistics,Sofia,Bulgaria,2013:243-248.
[5]Zhang Dong-yu,Yang Liang,Zheng Pu-qi,et al.Construction and application of affective metaphor corpus[J].Scientia Sinica(Informationis),2015,12(45):1574-1587.
[6]Lin Hong-fei,Zhang Dong-yu,Yang Liang,et al.Computational humor researches and application[J].Journal of Shandong University,2016,7(51):1-10.
[7]Dario Bertero,Pascale Fung.A long short-term memory framework for predicting humor in dialogues[C].Proceedings of North American Chapter of the Association for Computational Linguistics,San Diego,CA,2016:130-135.
[8]Zeng Xiao-bing,Zhang Zhi-ping,Liu Rong.Investigation and discussion on Chinese idioms and idiomatic phrases in the Chinese language situation report[J].Journal of Chinese Information Processing,2008,22(6):43-49.
[9]Xu Yao-min.Stereotypes,definitions and demarcation of idioms[J].Studies of the Chinese Language,1997,46(1):11-17.
[10]Ni Bao-yuan,Yao Peng-ci.Four questions of synonymous idiom[J].Studies of the Chinese Language,1995,44(1):23-28.
[11]Luo Ling,Chen Yi-dong,Shi Xiao-dong,et al.Chinese idiom translation based on paragraphing[J].Journal of Chinese Information Processing,2015,29(4):166-174.
[12]Ran Jie,Sun Yu,Chang Xia.Research of construction of the idiom story ontology based on OWL[J].Computer Technology and Development,2010,20(5):63-66.
[13]Yang Xue-song,Ran Jie.Research of semantic retrieval model based on idiom literary quotation ontology[J].Computer Science,2011,38(10A):219-221.
[14]Zhang Pu.Thinking about language sense and circulation[J].Language Teaching and Linguistic Studies,1999,21(2):83-96.
[15]Lv Shu-xiang.A preliminary study on the problem of single and double syllables in modern Chinese[J].Studies of the Chinese Language,1963,12(1):10-22.
[16]Li Xu.Humor and humor creation[J].Literary Review,1991,33(4):152-154.
[17]Fu Yun.The use of rhetoric in humor[J].Journal of Zhengzhou University,2005,38(4):155-156.
[18]Bu Zhan-ting.Study on the attitude of hyperbole[J].Contemporary Rhetoric,2010,29(4):53-59.
[19]Qiu Jin,Liao Qiao-yun.Hyperbole research under mind philosophy[J].Contemporary Rhetoric,2014,33(4):74-80.
[20]Zhu Yang.Specific case of humorous utterance translation from the perspective of relevance theory[J].Foreign Languages and Their Teaching,2007,29(2):57-60
[21]Xu Lin-hong,Lin Hong-fei,Pan Yu.Constructing the affective lexcion ontology[J].Journal of the China Society for Scientific and Technical Information,2008,27(2):180-185.
[22]Qian Kai-yu,Guo Li-peng.Research on sentiment analysis of social netw ork review s w ith idioms[J].Journal of Chinese Computer Systems,2017,6(38):1273-1277.
[23]Dictionary of the idiom[M].Beijing:The Commercial Press,2017.
[24]Zeng Jie-ying,Yu Bai-lin.The effects of Chinese characters'construction under thetachistoscopic repetition and non-tachistoscopic conditions[J].Psychological Science,1999,36(22):305-309.
[25]Gao Jia-ying.Modern Chinese calligraphy[M].Beijing:Higher Education Press,1993.
[26]Yu Bai-lin,Cao He-xi.A new exploration on the effect of strokenumber in identification of Chinese characters[J].Acta Psychologica Sinica,1992,37(2):120-126.
[27]Cao Chuan-yong,Shen Ye.The recognition of Chinese characters under tachistoscopic condition in primary school children[J].Acta Psychologica Sinica,1963,8(4):271-279.
[28]Hofmann K,Whiteson S,de Rijke M.Balancing exploration and exploitation in listwise and pairwise online learning to rank for information retrieval[J].Information Retrieval,2013,16(1):63-90.
[29]Wu Q,Burges C J C,Svore K M.Adapting boosting for information retrieval measures[J].Information Retrieval,2010,13(3):254-270.
[30]Liu Chang-zheng,Qin Peng.An investigation of idioms in Chinese new spapers based on the dynamic circulation corpus(DCC)[J].Applied Linguistics,2007,16(3):78-86.
[31]Zhang Tie-wen.The number and generation of idioms[J].Language Planning,1999,44(5):23-24.
[32]Xu Zhao-ben.Idiom knowledge introduction[M].Beijing:Bejing Press,1980.
[1]黄伯荣,廖序东.现代汉语[M].北京:高等教育出版社,2011.
[2]新华成语词典[M].北京:商务印书馆,2002.
[5]张冬瑜,杨亮,郑朴琪,等.情感隐喻语料库构建与应用[J].中国科学,2015,12(45):1574-1587.
[6]林鸿飞,张冬瑜,杨亮,等.幽默计算及其应用研究[J].山东大学学报,2016,7(51):1-10.
[8]曾小兵,张志平,刘荣.《中国语言生活状况报告》中成语与习语的调查与思考[J].中文信息学报,2008,22(6):43-49.
[9]徐耀民.成语的划界、定型和释义问题[J].中国语文,1997,46(1):11-17.
[10]倪宝元,姚鹏慈.等义成语四题[J].中国语文,1995,44(1):23-28.
[11]罗凌,陈毅东,史晓东,等.基于复述技术的汉语成语翻译方法研究[J].中文信息学报,2015,29(4):166-174.
[12]冉婕,孙瑜,昌霞.基于OWL的成语典故本体构建研究[J].计算机技术与发展,2010,20(5):63-66.
[13]杨雪松,冉婕.基于成语典故本体的语义检索模型研究[J].计算机科学,2011,38(10A):219-221.
[14]张普.关于语感与流通度的思考[J].语言教学与研究,1999,21(2):83-96.
[15]吕叔湘.现代汉语单双音节问题初探[J].中国语文,1963,12(1):10-22.
[16]李旭.幽默和幽默创造[J].文学评论,1991,33(4):152-154.
[17]付云.幽默中修辞手法的运用[J].郑州大学学报,2005,38(4):155-156.
[18]布占廷.夸张修辞的态度意义研究[J].当代修辞学,2010,29(4):53-59.
[19]邱晋,廖巧云.心智哲学视域下的夸张修辞研究[J].当代修辞学,2014,33(4):74-80.
[20]朱燕.关联理论对幽默言语翻译的诠释力[J].外语与外语教学,2007,29(2):57-60.
[21]徐琳宏,林鸿飞,潘宇.情感词汇本体的构造[J].情报学报,2008,27(2):180-185.
[22]钱凯雨,郭立鹏.融入习语信息的网络评论情感分析研究[J].小型微型计算机系统,2017,6(38):1273-1277.
[23]成语大词典[M].北京:商务印书馆,2017.
[24]曾捷英,喻柏林.速示重复和非速示呈现下的汉字结构方式效应[J].心理科学,1999,36(22):305-309.
[25]高家莺.现代汉字学[M].北京:高等教育出版社,1993.
[26]喻柏林,曹河析.汉字识别中的笔画数效应新探-兼论字频效应[J].心理学报,1992,37(2):120-126.
[27]曹传詠,沈晔.在速示条件下儿童辨认汉字字形的试探性研究[J].心理学报,1963,8(4):271-279.
[30]刘长征,秦鹏.基于中国主流报纸动态流通语料库(DCC)的成语使用情况调查[J].语言文字应用,2007,16(3):78-86.
[31]张铁文.成语的数量及产生年代[J].语文建设,1999,44(5):23-24.
[32]许肇本.成语知识浅谈[M].北京:北京出版社,1980.