基于机器学习算法的隐喻识别研究
|
摘要
隐喻是自然语言处理领域最棘手的问题之一,这个问题逐渐引起了学者们的关注,并且隐喻在思维及语言中所处的中心地位也逐渐为自然语言处理研究者所认同。隐喻是通过一个事物来表达另外一个事物,它不仅是一种修辞手段,而且体现着人的一种类比认知和思维方式。实际上,隐喻现象是一切自然语言中普遍存在的现象,是自然语言处理不可回避的问题,因此隐喻问题若得不到很好的解决,将成为制约自然语言处理和机器翻译发展的瓶颈。
近几年来,在隐喻识别方面,机器学习方法和大规模知识获取成了新的亮点。本文选择汉语文本中的隐喻计算问题作为研究对象,以隐喻识别为研究内容,从名词性隐喻和动词性隐喻这两种最主要的隐喻类型入手,采用多种机器学习算法,广泛地探讨了隐喻的识别方法。
本文选择了20个常用的隐喻词语,使用2001年—2004年的《人民日报》语料进行隐喻识别研究,主要研究内容如下：
基于有监督学习的隐喻识别。针对名词性隐喻和动词性隐喻这两类主要的隐喻类型,提出了基于RFR_SUM模型、SVM算法、CRF模型、最大熵模型和语义相似度计算模型的隐喻识别方法。分类方法为识别隐喻提供了一个机器识别的思路,考察了主流的分类方法在识别隐喻时的性能和效果。其中,RFR_SUM模型识别性能比较稳定,名词性和动词性隐喻的正确率的稳定性保持地最好。另外,CRF模型的识别正确率比SVM算法的略高,但是效果最好的是语义相似度计算模型。这个模型在采用K近邻分类算法的基础上,融入了语义相似度计算,使它的隐喻识别性能得到提高,是五个模型中隐喻识别正确率最高的。此外,对五个模型的实验结果进行了基于投票的集成实验。使隐喻识别正确率得到提升,名词性隐喻正确率达到87.74%,动词性隐喻正确率达到85.27%。
基于聚类算法的隐喻识别。在聚类过程中,样本间的相似度使用了基于同义词词林的向量空间相似度计算和基于知网的语义相似度计算两种方法,采用K-means算法原理,并对K-means算法的随机选择初始聚类中心的方法进行了优化。聚类实验设计了三个方案来提升隐喻识别结果的正确率,其中方案二不仅利用了近距离搭配信息,也考虑了远距离的特征信息,使实验结果得到提高。
基于半监督算法的隐喻识别。探讨了利用K-means算法和RFR SUM模型对隐喻识别进行半监督学习的方法,该算法是在分类实验和聚类实验的基础之上实现的。不仅利用了已标记样本的信息量,而且运用了未标记样本提供的信息,提高了隐喻识别的正确率。
最后,构建了用于隐喻计算的小型隐喻知识库。在隐喻研究的实验结果基础上,利用算法抽取隐喻类的特征词,并把特征词依据对应的RFR值进行排序,建立基于隐喻特征-RFR值这种结构的隐喻知识库。此外,通过基于知识库的隐喻计算实验,验证了构建的隐喻知识库的有效性。
总之,本文的研究工作主要是基于机器学习算法和知识获取方法,探讨了各种机器学习算法进行隐喻识别的实验思路,避免了手工知识库和规则方法的不足,积累了多种机器学习算法进行隐喻识别的大量实验数据,获得了隐喻识别研究较理想的实验结果。本文研究方法可以为隐喻计算、隐喻理解、隐喻本体研究及自然语言处理相关研究提供支持。
As one of the intractable problems in field of NLP(Natural Language Processing), metaphor has attracted more attention from researchers in recent years. And researchers have realized that it is the focus of mind and language mechanism. Metaphor is to express one thing in terms of another based on some similarities between the two things. It is not only a rhetorical devices of language, but also embodies people's analogical cognitive and way of thinking. In fact, the metaphor is prevalent phenomenon in all of the natural language. Also the metaphor problem can not be avoided in NLP field. So, if the problem is not well resolved, it will become a bottleneck of NLP and machine translation development.
In recent years, machine learning methods and automatic large-scale knowledge acquisition become popular in metaphor recognition. We select metaphor calculation as research subjects in Chinese text and metaphor recognition as research contents. In this thesis, we use many machine learning algorithms to study nominal metaphor and verbal metaphor and to explor wildly many methods of metaphor recognition.
The thesis chooses 20 metaphor words and uses the 2001 to 2004 the "People's Daily" Corpus to study metaphor recognition. The details are as follows:
Metaphor recognition based on classification algorithm. Basing on RFR_SUM, SVM, CRFs, maximum entropy and semantic similarity model based on How-Net, we present some recognition methods to process the problems of nominal metaphor recognition and verbal metaphor recognition. Classification algorithms provide an idea of machine recognition for metaphor recognition, so that we can study performance and effectiveness of mainstream classification models in identifying metaphor. The results show that the recognition performance of RFR_SUM model is relatively stable, because its recognition precision stability is best in the five models. In addition, CRF model recognition precision is slightly higher than SVM. But the best model is the semantic similarity, which combines semantic similarity calculation and the idea of K nearest neighbor algorithm, improving the metaphor recognition precision. Finally, based on observation of the experiment outcome of these models, an additional ensemble method based on majority voting is proposed. The ensemble method obtains nominal metaphor precision of 87.74% and verbal metaphor precision of 85.27%, which is much better than the results obtained in the five models.
Metaphor recognition based on clustering algorithm. In the clustering process, we use vector space similarity calculation based on TongYiCi CiLin and semantic similarity calculation based on How-Net to obtain the similarity between samples. Also we adopt the idea of K-means algorithm and optimize the mothed of selecting randomly initial cluster centers. Clustering experiments design three programs to enhance metaphor recognition precision, and the second experiment not only use short distance information but also long distance information, improving experimental results precision.
Metaphor recognition based on semi-supervised learning algorithm. We present semi-supervised learning method to metaphor recognition based on combining K-means algorithm and RFR_SUM model. This new method use both labeled samples set information and unlabeled samples set information, its prescion is higher than K-means clustering algorithm and RFR_SUM classification model.
Finally, we build a small metaphor knowledge-base for the metaphor computing. Based on the experimental results of metaphor study, we select feature words of metaphor class by using algorithm and sort these feature words by their RFR values, then establish our metaphor knowledge-base based on the structure of Feature-RFR. Furthermore, it is verified that the metaphor knowledge-base is available by metaphor computation experiments basing on our knowledge-base.
In short, the research contents in this thesis are mainly based on machine learning and knowledge acquisition, exploring experiment ideas of metaphor identification from some machine learning algorithms, avoiding shortages of mannual knowledge-bases and rule-based methods, accumulating much experimental data of machine learning algorithms in identifying metaphor, obtaining more satisfactory experiment results in metaphor recognition research. The research methods in this thesis can support researches on metaphor computing, metaphor understanding and other related natural language processing work.
近几年来,在隐喻识别方面,机器学习方法和大规模知识获取成了新的亮点。本文选择汉语文本中的隐喻计算问题作为研究对象,以隐喻识别为研究内容,从名词性隐喻和动词性隐喻这两种最主要的隐喻类型入手,采用多种机器学习算法,广泛地探讨了隐喻的识别方法。
本文选择了20个常用的隐喻词语,使用2001年—2004年的《人民日报》语料进行隐喻识别研究,主要研究内容如下：
基于有监督学习的隐喻识别。针对名词性隐喻和动词性隐喻这两类主要的隐喻类型,提出了基于RFR_SUM模型、SVM算法、CRF模型、最大熵模型和语义相似度计算模型的隐喻识别方法。分类方法为识别隐喻提供了一个机器识别的思路,考察了主流的分类方法在识别隐喻时的性能和效果。其中,RFR_SUM模型识别性能比较稳定,名词性和动词性隐喻的正确率的稳定性保持地最好。另外,CRF模型的识别正确率比SVM算法的略高,但是效果最好的是语义相似度计算模型。这个模型在采用K近邻分类算法的基础上,融入了语义相似度计算,使它的隐喻识别性能得到提高,是五个模型中隐喻识别正确率最高的。此外,对五个模型的实验结果进行了基于投票的集成实验。使隐喻识别正确率得到提升,名词性隐喻正确率达到87.74%,动词性隐喻正确率达到85.27%。
基于聚类算法的隐喻识别。在聚类过程中,样本间的相似度使用了基于同义词词林的向量空间相似度计算和基于知网的语义相似度计算两种方法,采用K-means算法原理,并对K-means算法的随机选择初始聚类中心的方法进行了优化。聚类实验设计了三个方案来提升隐喻识别结果的正确率,其中方案二不仅利用了近距离搭配信息,也考虑了远距离的特征信息,使实验结果得到提高。
基于半监督算法的隐喻识别。探讨了利用K-means算法和RFR SUM模型对隐喻识别进行半监督学习的方法,该算法是在分类实验和聚类实验的基础之上实现的。不仅利用了已标记样本的信息量,而且运用了未标记样本提供的信息,提高了隐喻识别的正确率。
最后,构建了用于隐喻计算的小型隐喻知识库。在隐喻研究的实验结果基础上,利用算法抽取隐喻类的特征词,并把特征词依据对应的RFR值进行排序,建立基于隐喻特征-RFR值这种结构的隐喻知识库。此外,通过基于知识库的隐喻计算实验,验证了构建的隐喻知识库的有效性。
总之,本文的研究工作主要是基于机器学习算法和知识获取方法,探讨了各种机器学习算法进行隐喻识别的实验思路,避免了手工知识库和规则方法的不足,积累了多种机器学习算法进行隐喻识别的大量实验数据,获得了隐喻识别研究较理想的实验结果。本文研究方法可以为隐喻计算、隐喻理解、隐喻本体研究及自然语言处理相关研究提供支持。
As one of the intractable problems in field of NLP(Natural Language Processing), metaphor has attracted more attention from researchers in recent years. And researchers have realized that it is the focus of mind and language mechanism. Metaphor is to express one thing in terms of another based on some similarities between the two things. It is not only a rhetorical devices of language, but also embodies people's analogical cognitive and way of thinking. In fact, the metaphor is prevalent phenomenon in all of the natural language. Also the metaphor problem can not be avoided in NLP field. So, if the problem is not well resolved, it will become a bottleneck of NLP and machine translation development.
In recent years, machine learning methods and automatic large-scale knowledge acquisition become popular in metaphor recognition. We select metaphor calculation as research subjects in Chinese text and metaphor recognition as research contents. In this thesis, we use many machine learning algorithms to study nominal metaphor and verbal metaphor and to explor wildly many methods of metaphor recognition.
The thesis chooses 20 metaphor words and uses the 2001 to 2004 the "People's Daily" Corpus to study metaphor recognition. The details are as follows:
Metaphor recognition based on classification algorithm. Basing on RFR_SUM, SVM, CRFs, maximum entropy and semantic similarity model based on How-Net, we present some recognition methods to process the problems of nominal metaphor recognition and verbal metaphor recognition. Classification algorithms provide an idea of machine recognition for metaphor recognition, so that we can study performance and effectiveness of mainstream classification models in identifying metaphor. The results show that the recognition performance of RFR_SUM model is relatively stable, because its recognition precision stability is best in the five models. In addition, CRF model recognition precision is slightly higher than SVM. But the best model is the semantic similarity, which combines semantic similarity calculation and the idea of K nearest neighbor algorithm, improving the metaphor recognition precision. Finally, based on observation of the experiment outcome of these models, an additional ensemble method based on majority voting is proposed. The ensemble method obtains nominal metaphor precision of 87.74% and verbal metaphor precision of 85.27%, which is much better than the results obtained in the five models.
Metaphor recognition based on clustering algorithm. In the clustering process, we use vector space similarity calculation based on TongYiCi CiLin and semantic similarity calculation based on How-Net to obtain the similarity between samples. Also we adopt the idea of K-means algorithm and optimize the mothed of selecting randomly initial cluster centers. Clustering experiments design three programs to enhance metaphor recognition precision, and the second experiment not only use short distance information but also long distance information, improving experimental results precision.
Metaphor recognition based on semi-supervised learning algorithm. We present semi-supervised learning method to metaphor recognition based on combining K-means algorithm and RFR_SUM model. This new method use both labeled samples set information and unlabeled samples set information, its prescion is higher than K-means clustering algorithm and RFR_SUM classification model.
Finally, we build a small metaphor knowledge-base for the metaphor computing. Based on the experimental results of metaphor study, we select feature words of metaphor class by using algorithm and sort these feature words by their RFR values, then establish our metaphor knowledge-base based on the structure of Feature-RFR. Furthermore, it is verified that the metaphor knowledge-base is available by metaphor computation experiments basing on our knowledge-base.
In short, the research contents in this thesis are mainly based on machine learning and knowledge acquisition, exploring experiment ideas of metaphor identification from some machine learning algorithms, avoiding shortages of mannual knowledge-bases and rule-based methods, accumulating much experimental data of machine learning algorithms in identifying metaphor, obtaining more satisfactory experiment results in metaphor recognition research. The research methods in this thesis can support researches on metaphor computing, metaphor understanding and other related natural language processing work.
引文
[1]Lakoff, G., Johnson, M. Metaphors We Live By. Chicago:University of Chicago Press,1980.
[2]杨芸.汉语隐喻识别与解释计算模型研究[D].厦门大学博士学位论文,2008.
[3]Lakoff, G. The contemporary theory of metaphor//Ortony A. Metaphor and thought.2ed. Cambridge:Cambridge University Press,1993:202-251.
[4]贾玉祥.汉语文本中的隐喻计算研究[D].北京大学博士学位论文,2010.
[5]王治敏.汉语名词短语隐喻识别研究[D].北京大学博士学位论文,2006.
[6]唐旭日.谓词语义计算：搭配与概念隐喻[D].南京师范大学博士学位论文,2010.
[7]Fass, D., met*:A Method for Discriminating Metonymy and Metaphor by Computer[C]. Association for Computational Linguistics.1991.17. Numberl.
[8]Resnik, P. Using Information Content to Evaluate Semantic Similarity in a Taxonomy [C]. The 14th International Joint Conference on Artificial Intelligent, Montreal. Canada.1995.
[9]Weiner, E. J. A Knowledge Representation Approach to Understanding Metaphors[J]. Computational Linguistics,1984,10(1).
[10]Martin, J. H. A Computational Model of Metaphor Interpretation[M]. Academic Press, San Diego, CA,1990.
[11]Veale, T. Metaphor, Memory and Meaning:Symbolic and Connecionist Issues in Metaphor Interpretation[D]. PhD thesis in Trinity College, Dublin,1995.
[12]Steinhart, E.C. The Logic of Metaphor:Analogous Parts of Possible Worlds[M]. Dordrecht:Kluwer Academic Publishers,2001.
[13]Gentner, D. Structure-Mapping:A Theoretical Framework for Analogy[J]. Cognitive Science,1983,7.
[14]Gentner, D. Falkenhainer, B., and Skorstad, J. Viewing Metaphor as analogy[C]. In Helman, D.,editor, Analogical Reasoning. Kluwer Academic Publishers,1988.
[15]Holyoak, K. J., Thagard, P. Analogical Mapping by Constraint Satisfaction[J]. Cognitive Science.1989, Vol.13.
[16]Kinstch, W. Metaphor Comprehension:A computational Theory[J]. Psychonomic Bulletin and Review,2000.
[17]Kinstch, W., Bowles, A. Metaphor comprehension:What makes a metaphor difficult to understand?[J]. Metaphor and Symbol.2002, Vol.17.
[18]Mason, Z. Computational, Corpus-based Metaphor Extraction System[D]. Brandeis University,2002.
[19]Mason, Z. CorMet:A Computational, Corpus-based Conventional Metaphor Extraction System[J]. Computational Linguistics,2004,30(1).
[20]俞士汶.语料库与综合型语言知识库的建设[C].中文信息处理若干重要问题,2002.
[21]Zhimin Wang, Houfeng Wang, Huiming Duan, Shuang Han, Shifen Yu. Chinese Noun Phrase Metaphor Recognition with Maximum Entropy Approach[C]. In:Proceedings of the Seventh International Conference on Intelligent Text Processing and Computational Linguistics. Mexico,2006.
[22]周昌乐.隐喻、类比逻辑与可能世界[J].外国语言文学研究.2004,4(6).
[23]周昌乐.认知逻辑导论[M].北京：清华大学出版社,2001.
[24]张威,周昌乐.汉语隐喻理解的逻辑描述初探[J].中文信息学报,2004,18 (5)
[25]杨芸,周昌乐.基于机器理解的汉语隐喻分类研究初步[J].中文信息学报,2004,18(4)
[26]戴帅湘.汉语隐喻分类识别的计算方法研究[D].厦门大学硕士学位论文,2005,6.
[27]王雪梅.基于机器理解的汉语比喻再归类研究[D].厦门大学硕士学位论文,2005,6.
[28]黄孝喜,周昌乐.隐喻理解的计算模型综述[J].计算机科学.2006,33(8).
[29]Xiaoxi Huang, Yun Yang and Changle Zhou. Emotional Metaphors for Emotion Recognition in Chinese Text[C]. In:Proceedings of First International Conference Affective Computing and Intelligence Interaction, LNCS 3784.
[30]袁毓林.容器隐喻、套件隐喻及相关的语法现象---词语同现限制的认知解释和计算分析[J].中国语文,2004(3)
[31]Pragglejaz Group. MIP:A Method for Identifying Metaphorically Used Words in Discourse. Metaphor and Symbol,2007,22(1):1-39.
[32]束定芳.隐喻学研究[M].上海：上海外语教育出版社,2000.
[33]Wallington, A. M. et al. Metaphoricity Signals:A Corpus-Based Investigation. Technical Report CSRP-03-05, University of Birmingham,2003.
[34]Ferrari, S. Using Textual Clues to Improve Metaphor Processing. In Proceedings of the 34th Annual Meeting on Association for Computational Linguistics Student Session, Santa Cruz, California,1996,351-353.
[35]Wilks, Y. A Preferential Pattern-seeking Semantics for Natural Language Inference. Artificial Intelligence,1975,6:53-74.
[36]Fass, D. Met*. A Method for Discriminating Metonymy and Metaphor By Computer. Computational Linguistics,1991,17(1):49-90.
[37]Mason, Z. CorMet:A Computational, Corpus-based Conventional Metaphor Extraction System. Computational Linguistics,2004,30(1):23-44.
[38]Krishnakumaran, S., Zhu, X. J. Hunting Elusive Metaphors Using Lexical Resources. In Proceedings of the ACL 2007 Workshop on Computational Approaches to Figurative Language,2007,13-20.
[39]Brants, T. and Franz, A. Web 1T 5-gram Version 1. Linguistic Data Consortium, Philadelphia,2006.
[40]董振东,董强.How-Net.http://www.keenage.com.
[41]Gedigian, M., Bryant, J., Narayannan,S., Ciric, B. Catching Metaphors. In Proceedings of the 3rd Workshop on Scalable Natural Language Understanding, 2006,41-48.
[42]Birke, J., Sarkar, A. A Clustering Approach for the Nearly Unsupervised Recognition of Nonliteral Language. In Proceedings of the 11th Conference of the European Chapter of the Association for Computational Linguistics. Trento Italy,2006,329-336.
[43]Karov, Y., Edelman, S. Similarity-Based Word Sense Disambiguation. Computational Linguistics,1998,24(1):41-59.
[44]Pasanek, B., Sculley, D. Mining Millions of Metaphors. Literary and Linguistic Computing,2006,23(3):345-360.
[45]Steven Bethard, Vicky Tzuyin Lai, James, Martin, H. Topic Model Analysis of Metaphor Frequency for Psycholinguistic Stimuli. In Proceedings of the NAACL HLT Workshop on Computational Approaches to Linguistic Creativity, Boulder, Colorado, June 2009,9-16.
[46]曲维光.现代汉语词语级歧义自动消解研究[M].北京：科学出版社,2008.
[47]邓乃扬,田英杰.支持向量机——理论、算法与拓展[M].北京：科学出版社, 2009：79-111.
[48]John Lafferty, Andrew McCallum, et al. Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data[A]. Proc. International Conference on Machine Learning[C], San Francisco:Mogan Koufmann,2001:282-289.
[49]Chen Wen liang, Zhang Yu-jie, et al. Chinese Named Entity Recognition with Conditional Random Fields.Proceedings of the Fifth SIGHAN Workshop on Chinese Language Processing,2006.
[50]Adwait R. A simple introduction to maximum entropy models for natural language process[R]. Philadelphia, PA, USA:University of Pennsylvania, Tech. Rep:IRCS-97-08,1997.
[51]Cover, T. M. Hart, P. E. Nearest neighbor pattern classification[J]. IEEE Transaction of informationTheory,1967,13(1):21-27.
[52]刘群,李素建.基于《知网》的词汇语义相似度的计算[A].第三届汉语词汇语义学研讨会[C],2002.
[53]丁德鑫,曲维光,于丽丽等.基于词频和语义信息的组合型歧义消解[A].中国计算语言学研究前沿进展——第十届全国计算语言学学术会议论文集[C],2009.
[54]Dietterich, T. G. Ensemble Methods in Machine Learning. In Multiple Classier Systems, Cagliari, Italy,2000.
[55]Valentini, G. and Masulli, F. "Ensembles of learning machines," in Neural Nets WIRN Vietri-02, Series Lecture Notes in Computer Sciences, M. Marinaro And R. Tagliaferri, Eds.:Springer-Verlag, Heidelberg (Germany),2002, Invited Review.
[56]Thomas, G. Dietterich. Ensemble learning. In The Handbook of Brain Theory and Neural Networks, Second Edition,2002.
[57]梁英毅.集成学习综述.http://soft.cs.tsinghua.edu.cn/-keltin/do cs/ensemble.pdf.
[58]ShiXin Yu. Feature Selection and Classifier Ensembles:A Study on yperspectral Remote Sensing Data,2003. Available at http://143.129.203.3/visielab/theses/shixin/thesis_yu.pdf.
[59]Wolpert, D. H. Stacked Generalization, Neural Networks,Pergamon Press,1992, (5),241-259.
[60]Ricardo V., Youssef, D.:A perspective view and survey of meta-learning. Artificial Intelligence Review,2002,18(2):77-95.
[61]刘鑫朝.聚类算法的研究及其应用[D].长沙理工大学硕士学位论文,2006.
[62]Anil, K. Jain, R. Statistical pattern recognition:A review [J], IEEE Transactions on pattern analysis and machine intelligence.2000,22(1):1-37.
[63]Manning, C. D., Schutze, H. Foundations of statistican natural language processing[M]. MA:MIT Press,1999.
[64]徐涛.词语相似度计算方法研究[D].南京师范大学硕士学位论文,2010.
[65]万星火,檀亦丽.数据挖掘的聚类方法[J].统计与决策,2005.27(18)：125-126.
[66]易星.半监督学习若干问题的研究[D].清华大学硕士学位论文,2004.
[67]Rosset, S., Zhu, J., Zou, H. et al. A method for inferring label sampling mechanisms in semi-supervised learning[C]. In:Saul L K, Weiss Y, Bottou L, eds. Advances in Neural Information Processing Systems 17. Cambridge, MA: MIT Press,2005.1161-1168.
[68]Liu, B., Lee, W. S., Yu, P. S. et al. Partially supervised classification of text documents[C]. In:Sammut C, Hoffmann A G, eds. Proceedings of the Nineteenth International Conference on Machine Learning.San Francisco, CA: Morgan Kaufmann,2002.387-394.
[69]Klein, D., Kamvar, S. D. and Manning, C. From instance-level costraints to space-level constraints:making the most of prior knowledge in data clustering. In Proceedings of the 9th International Conference on Machine Learning.2002. Sydney, Australia.
[70]周志华.半监督学习中的协同训练算法[M]//周志华,王珏.机器学习及其应用.北京：清华大学出版社,2007.
[71]Shahshahani, B., Landgrebe, D. The Effect of Unlabeled Samples in Reducing the Sma 11 Sample Size Pr oblem and Mitigating the H ughes Phenomenon [J]. IE EE T r ansactions on Geoscience a nd Remote Sensing,1994,32 (5).
[72]梁吉业,高嘉伟,常瑜.半监督学习研究进展[J].山西大学学报(自然科学版)32(4)：528-534,2009.
[73]Bensaid, A. M., Hall, L. O., Bezdek, J. C. et al. Partially supervised clustering for image segmentation[J]. Pattern Recognition,1996,29(5):859-871.
[74]Miller, D. J., Uyar, H. S. A mixture of experts classifier with learning based on both labelled and unlabelled data[C]. In:Mozer M, Jordan M I, Petsche T, eds. Advances in Neural Information Processing Systems 9. Cambridge, MA:MIT Press 1997.571-577.
[75]Dempster, A. P., Laird, N. M., Rubin, D. B. Maximum likelihood from incomplete data via the EM algorithm[J]. Journal of the Royal Statistical Society, Series B (Methodological),1977,39(1):1-38.
[76]Seeger, M. Learning with labeled and unlabeled data[R]. Technical Report, Institute for Adaptive and Neural Computation,University of Edinburgh,2001. 1-7.
[77]孔怡青.半监督学习及其应用研究[D].江南大学博士学位论文,2009.
[78]Wang, F., Wang, J. D., Zhang, C. S. et al. Semi-Supervised classification using linear neighborhood propagation[C]. Proc. Of IEEE Conference on Computer Vision and Pattern Recognition,2006, pp.160-167.
[79]Bennett, K. Combing support vector and mathematical programming methods for classification[M]. Advances in Kernel Methods-Support Vector Learning, MIT Press,1998:234-301.
[80]张霄军,曲维光.国内外隐喻知识库建设综述[J].计算机应用研究,2008.
[81]王金锦.面向隐喻计算的实体概念知识库构建方法研究[D].厦门大学硕士学位论文,2009.
[82]李剑锋,杨芸,周昌乐.面向隐喻计算的语料库研究和建设[J].心智与计算,2007(1)：142-146.
[83]尤昉,李涓子,王作英.基于语义依存关系的汉语语料库的构建[J].中文信息学报,2002,17(1)：46-53.
[2]杨芸.汉语隐喻识别与解释计算模型研究[D].厦门大学博士学位论文,2008.
[3]Lakoff, G. The contemporary theory of metaphor//Ortony A. Metaphor and thought.2ed. Cambridge:Cambridge University Press,1993:202-251.
[4]贾玉祥.汉语文本中的隐喻计算研究[D].北京大学博士学位论文,2010.
[5]王治敏.汉语名词短语隐喻识别研究[D].北京大学博士学位论文,2006.
[6]唐旭日.谓词语义计算：搭配与概念隐喻[D].南京师范大学博士学位论文,2010.
[7]Fass, D., met*:A Method for Discriminating Metonymy and Metaphor by Computer[C]. Association for Computational Linguistics.1991.17. Numberl.
[8]Resnik, P. Using Information Content to Evaluate Semantic Similarity in a Taxonomy [C]. The 14th International Joint Conference on Artificial Intelligent, Montreal. Canada.1995.
[9]Weiner, E. J. A Knowledge Representation Approach to Understanding Metaphors[J]. Computational Linguistics,1984,10(1).
[10]Martin, J. H. A Computational Model of Metaphor Interpretation[M]. Academic Press, San Diego, CA,1990.
[11]Veale, T. Metaphor, Memory and Meaning:Symbolic and Connecionist Issues in Metaphor Interpretation[D]. PhD thesis in Trinity College, Dublin,1995.
[12]Steinhart, E.C. The Logic of Metaphor:Analogous Parts of Possible Worlds[M]. Dordrecht:Kluwer Academic Publishers,2001.
[13]Gentner, D. Structure-Mapping:A Theoretical Framework for Analogy[J]. Cognitive Science,1983,7.
[14]Gentner, D. Falkenhainer, B., and Skorstad, J. Viewing Metaphor as analogy[C]. In Helman, D.,editor, Analogical Reasoning. Kluwer Academic Publishers,1988.
[15]Holyoak, K. J., Thagard, P. Analogical Mapping by Constraint Satisfaction[J]. Cognitive Science.1989, Vol.13.
[16]Kinstch, W. Metaphor Comprehension:A computational Theory[J]. Psychonomic Bulletin and Review,2000.
[17]Kinstch, W., Bowles, A. Metaphor comprehension:What makes a metaphor difficult to understand?[J]. Metaphor and Symbol.2002, Vol.17.
[18]Mason, Z. Computational, Corpus-based Metaphor Extraction System[D]. Brandeis University,2002.
[19]Mason, Z. CorMet:A Computational, Corpus-based Conventional Metaphor Extraction System[J]. Computational Linguistics,2004,30(1).
[20]俞士汶.语料库与综合型语言知识库的建设[C].中文信息处理若干重要问题,2002.
[21]Zhimin Wang, Houfeng Wang, Huiming Duan, Shuang Han, Shifen Yu. Chinese Noun Phrase Metaphor Recognition with Maximum Entropy Approach[C]. In:Proceedings of the Seventh International Conference on Intelligent Text Processing and Computational Linguistics. Mexico,2006.
[22]周昌乐.隐喻、类比逻辑与可能世界[J].外国语言文学研究.2004,4(6).
[23]周昌乐.认知逻辑导论[M].北京：清华大学出版社,2001.
[24]张威,周昌乐.汉语隐喻理解的逻辑描述初探[J].中文信息学报,2004,18 (5)
[25]杨芸,周昌乐.基于机器理解的汉语隐喻分类研究初步[J].中文信息学报,2004,18(4)
[26]戴帅湘.汉语隐喻分类识别的计算方法研究[D].厦门大学硕士学位论文,2005,6.
[27]王雪梅.基于机器理解的汉语比喻再归类研究[D].厦门大学硕士学位论文,2005,6.
[28]黄孝喜,周昌乐.隐喻理解的计算模型综述[J].计算机科学.2006,33(8).
[29]Xiaoxi Huang, Yun Yang and Changle Zhou. Emotional Metaphors for Emotion Recognition in Chinese Text[C]. In:Proceedings of First International Conference Affective Computing and Intelligence Interaction, LNCS 3784.
[30]袁毓林.容器隐喻、套件隐喻及相关的语法现象---词语同现限制的认知解释和计算分析[J].中国语文,2004(3)
[31]Pragglejaz Group. MIP:A Method for Identifying Metaphorically Used Words in Discourse. Metaphor and Symbol,2007,22(1):1-39.
[32]束定芳.隐喻学研究[M].上海：上海外语教育出版社,2000.
[33]Wallington, A. M. et al. Metaphoricity Signals:A Corpus-Based Investigation. Technical Report CSRP-03-05, University of Birmingham,2003.
[34]Ferrari, S. Using Textual Clues to Improve Metaphor Processing. In Proceedings of the 34th Annual Meeting on Association for Computational Linguistics Student Session, Santa Cruz, California,1996,351-353.
[35]Wilks, Y. A Preferential Pattern-seeking Semantics for Natural Language Inference. Artificial Intelligence,1975,6:53-74.
[36]Fass, D. Met*. A Method for Discriminating Metonymy and Metaphor By Computer. Computational Linguistics,1991,17(1):49-90.
[37]Mason, Z. CorMet:A Computational, Corpus-based Conventional Metaphor Extraction System. Computational Linguistics,2004,30(1):23-44.
[38]Krishnakumaran, S., Zhu, X. J. Hunting Elusive Metaphors Using Lexical Resources. In Proceedings of the ACL 2007 Workshop on Computational Approaches to Figurative Language,2007,13-20.
[39]Brants, T. and Franz, A. Web 1T 5-gram Version 1. Linguistic Data Consortium, Philadelphia,2006.
[40]董振东,董强.How-Net.http://www.keenage.com.
[41]Gedigian, M., Bryant, J., Narayannan,S., Ciric, B. Catching Metaphors. In Proceedings of the 3rd Workshop on Scalable Natural Language Understanding, 2006,41-48.
[42]Birke, J., Sarkar, A. A Clustering Approach for the Nearly Unsupervised Recognition of Nonliteral Language. In Proceedings of the 11th Conference of the European Chapter of the Association for Computational Linguistics. Trento Italy,2006,329-336.
[43]Karov, Y., Edelman, S. Similarity-Based Word Sense Disambiguation. Computational Linguistics,1998,24(1):41-59.
[44]Pasanek, B., Sculley, D. Mining Millions of Metaphors. Literary and Linguistic Computing,2006,23(3):345-360.
[45]Steven Bethard, Vicky Tzuyin Lai, James, Martin, H. Topic Model Analysis of Metaphor Frequency for Psycholinguistic Stimuli. In Proceedings of the NAACL HLT Workshop on Computational Approaches to Linguistic Creativity, Boulder, Colorado, June 2009,9-16.
[46]曲维光.现代汉语词语级歧义自动消解研究[M].北京：科学出版社,2008.
[47]邓乃扬,田英杰.支持向量机——理论、算法与拓展[M].北京：科学出版社, 2009：79-111.
[48]John Lafferty, Andrew McCallum, et al. Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data[A]. Proc. International Conference on Machine Learning[C], San Francisco:Mogan Koufmann,2001:282-289.
[49]Chen Wen liang, Zhang Yu-jie, et al. Chinese Named Entity Recognition with Conditional Random Fields.Proceedings of the Fifth SIGHAN Workshop on Chinese Language Processing,2006.
[50]Adwait R. A simple introduction to maximum entropy models for natural language process[R]. Philadelphia, PA, USA:University of Pennsylvania, Tech. Rep:IRCS-97-08,1997.
[51]Cover, T. M. Hart, P. E. Nearest neighbor pattern classification[J]. IEEE Transaction of informationTheory,1967,13(1):21-27.
[52]刘群,李素建.基于《知网》的词汇语义相似度的计算[A].第三届汉语词汇语义学研讨会[C],2002.
[53]丁德鑫,曲维光,于丽丽等.基于词频和语义信息的组合型歧义消解[A].中国计算语言学研究前沿进展——第十届全国计算语言学学术会议论文集[C],2009.
[54]Dietterich, T. G. Ensemble Methods in Machine Learning. In Multiple Classier Systems, Cagliari, Italy,2000.
[55]Valentini, G. and Masulli, F. "Ensembles of learning machines," in Neural Nets WIRN Vietri-02, Series Lecture Notes in Computer Sciences, M. Marinaro And R. Tagliaferri, Eds.:Springer-Verlag, Heidelberg (Germany),2002, Invited Review.
[56]Thomas, G. Dietterich. Ensemble learning. In The Handbook of Brain Theory and Neural Networks, Second Edition,2002.
[57]梁英毅.集成学习综述.http://soft.cs.tsinghua.edu.cn/-keltin/do cs/ensemble.pdf.
[58]ShiXin Yu. Feature Selection and Classifier Ensembles:A Study on yperspectral Remote Sensing Data,2003. Available at http://143.129.203.3/visielab/theses/shixin/thesis_yu.pdf.
[59]Wolpert, D. H. Stacked Generalization, Neural Networks,Pergamon Press,1992, (5),241-259.
[60]Ricardo V., Youssef, D.:A perspective view and survey of meta-learning. Artificial Intelligence Review,2002,18(2):77-95.
[61]刘鑫朝.聚类算法的研究及其应用[D].长沙理工大学硕士学位论文,2006.
[62]Anil, K. Jain, R. Statistical pattern recognition:A review [J], IEEE Transactions on pattern analysis and machine intelligence.2000,22(1):1-37.
[63]Manning, C. D., Schutze, H. Foundations of statistican natural language processing[M]. MA:MIT Press,1999.
[64]徐涛.词语相似度计算方法研究[D].南京师范大学硕士学位论文,2010.
[65]万星火,檀亦丽.数据挖掘的聚类方法[J].统计与决策,2005.27(18)：125-126.
[66]易星.半监督学习若干问题的研究[D].清华大学硕士学位论文,2004.
[67]Rosset, S., Zhu, J., Zou, H. et al. A method for inferring label sampling mechanisms in semi-supervised learning[C]. In:Saul L K, Weiss Y, Bottou L, eds. Advances in Neural Information Processing Systems 17. Cambridge, MA: MIT Press,2005.1161-1168.
[68]Liu, B., Lee, W. S., Yu, P. S. et al. Partially supervised classification of text documents[C]. In:Sammut C, Hoffmann A G, eds. Proceedings of the Nineteenth International Conference on Machine Learning.San Francisco, CA: Morgan Kaufmann,2002.387-394.
[69]Klein, D., Kamvar, S. D. and Manning, C. From instance-level costraints to space-level constraints:making the most of prior knowledge in data clustering. In Proceedings of the 9th International Conference on Machine Learning.2002. Sydney, Australia.
[70]周志华.半监督学习中的协同训练算法[M]//周志华,王珏.机器学习及其应用.北京：清华大学出版社,2007.
[71]Shahshahani, B., Landgrebe, D. The Effect of Unlabeled Samples in Reducing the Sma 11 Sample Size Pr oblem and Mitigating the H ughes Phenomenon [J]. IE EE T r ansactions on Geoscience a nd Remote Sensing,1994,32 (5).
[72]梁吉业,高嘉伟,常瑜.半监督学习研究进展[J].山西大学学报(自然科学版)32(4)：528-534,2009.
[73]Bensaid, A. M., Hall, L. O., Bezdek, J. C. et al. Partially supervised clustering for image segmentation[J]. Pattern Recognition,1996,29(5):859-871.
[74]Miller, D. J., Uyar, H. S. A mixture of experts classifier with learning based on both labelled and unlabelled data[C]. In:Mozer M, Jordan M I, Petsche T, eds. Advances in Neural Information Processing Systems 9. Cambridge, MA:MIT Press 1997.571-577.
[75]Dempster, A. P., Laird, N. M., Rubin, D. B. Maximum likelihood from incomplete data via the EM algorithm[J]. Journal of the Royal Statistical Society, Series B (Methodological),1977,39(1):1-38.
[76]Seeger, M. Learning with labeled and unlabeled data[R]. Technical Report, Institute for Adaptive and Neural Computation,University of Edinburgh,2001. 1-7.
[77]孔怡青.半监督学习及其应用研究[D].江南大学博士学位论文,2009.
[78]Wang, F., Wang, J. D., Zhang, C. S. et al. Semi-Supervised classification using linear neighborhood propagation[C]. Proc. Of IEEE Conference on Computer Vision and Pattern Recognition,2006, pp.160-167.
[79]Bennett, K. Combing support vector and mathematical programming methods for classification[M]. Advances in Kernel Methods-Support Vector Learning, MIT Press,1998:234-301.
[80]张霄军,曲维光.国内外隐喻知识库建设综述[J].计算机应用研究,2008.
[81]王金锦.面向隐喻计算的实体概念知识库构建方法研究[D].厦门大学硕士学位论文,2009.
[82]李剑锋,杨芸,周昌乐.面向隐喻计算的语料库研究和建设[J].心智与计算,2007(1)：142-146.
[83]尤昉,李涓子,王作英.基于语义依存关系的汉语语料库的构建[J].中文信息学报,2002,17(1)：46-53.