自然语言理解中代词语义分析及在产品设计中的应用
|
摘要
本文针对目前计算机辅助设计技术对产品设计的早期阶段支持不充分的问题,将自然语言理解应用于产品设计中,对以自然语言形式表达的用户需求中的代词进行理解和分析,同时结合系统其余模块的运行,将最终的分析结果转化成概念设计要求,为后续设计提供支持。
首先,根据设计领域自然语言理解的特点,确定应用于产品设计的自然语言理解应该采用基于知识为主的方法。其次,通过对目前现有的几种语义分析方法的分析和比较,选择本体理论作为基于知识的方法的重要补充,深入探讨了其在基于知识系统的优点。同时资源模型可以很方便地表示概念的内涵和外延,使用它可以对基于知识的方法起到辅助作用。利用基于知识的方法、本体理论和资源模型,建立了语义分析的知识表示体系。然后,根据本体理论建立了概念从属树来表示概念间的关系。在对汉语代词语义分析的基础上,用基于知识的方法建立了代词、动态代词模板,实现了汉语代词的自然语言理解。再次,在以上工作的基础上,编写出对汉语代词进行语义分析的程序,可以实现对领域内代词的理解。最后,将自然语言理解的语义分析应用于凸轮机构设计需求分析原型系统,对其中的代词进行识别,并结合系统其余模块的运行,经过初步调试,取得了一定的成果。
Presently the auxiliary designing technique of the computer designs for products mainly aimed at later stage, the support to its early stage is not enough. Based on the application of natural language understanding to product design, the Chinese Pronoun subordinate to user requirements, which expressed in natural language style, is understood and analyzed. With the whole system running, the results are conversed into the requirement of conceptual design to support the future design.
First, method based on knowledge should be adopted in the natural language understanding applied to product design domain according to the characteristic of natural language understanding in product design domain. Second, Ontology theory is chosen as an important complementarity to the method based on knowledge, discussed its strong suit in the system based on knowledge, through analysis and compares of several existing semantic analysis method present. Simultaneity, resource model can expediently express the concept connotation and the concept extension, it appears the assistant function of the method based on knowledge. The knowledge expression system of semantic analysis is set up on method based on knowledge, ontology and resource model. Conceptual dependency trees are set up based on the ontology theory to denote the relation of concepts. By detailed analysis of the intension and the extension of concepts, pronoun and dynamic pronoun classes templates are established on method based on knowledge. Accomplished the understanding of Chinese pronoun. Forth, Semantic analysis program is written to understand sentence on the basis of the work done, which can understand Chinese pronoun in corresponding fields. Last, Semantic analysis of natural language understanding is applied to requirement analysis archetypal system for the design of cam machine. The pronoun in user requirements is identified. With other models running, the system runs well after debug.
首先,根据设计领域自然语言理解的特点,确定应用于产品设计的自然语言理解应该采用基于知识为主的方法。其次,通过对目前现有的几种语义分析方法的分析和比较,选择本体理论作为基于知识的方法的重要补充,深入探讨了其在基于知识系统的优点。同时资源模型可以很方便地表示概念的内涵和外延,使用它可以对基于知识的方法起到辅助作用。利用基于知识的方法、本体理论和资源模型,建立了语义分析的知识表示体系。然后,根据本体理论建立了概念从属树来表示概念间的关系。在对汉语代词语义分析的基础上,用基于知识的方法建立了代词、动态代词模板,实现了汉语代词的自然语言理解。再次,在以上工作的基础上,编写出对汉语代词进行语义分析的程序,可以实现对领域内代词的理解。最后,将自然语言理解的语义分析应用于凸轮机构设计需求分析原型系统,对其中的代词进行识别,并结合系统其余模块的运行,经过初步调试,取得了一定的成果。
Presently the auxiliary designing technique of the computer designs for products mainly aimed at later stage, the support to its early stage is not enough. Based on the application of natural language understanding to product design, the Chinese Pronoun subordinate to user requirements, which expressed in natural language style, is understood and analyzed. With the whole system running, the results are conversed into the requirement of conceptual design to support the future design.
First, method based on knowledge should be adopted in the natural language understanding applied to product design domain according to the characteristic of natural language understanding in product design domain. Second, Ontology theory is chosen as an important complementarity to the method based on knowledge, discussed its strong suit in the system based on knowledge, through analysis and compares of several existing semantic analysis method present. Simultaneity, resource model can expediently express the concept connotation and the concept extension, it appears the assistant function of the method based on knowledge. The knowledge expression system of semantic analysis is set up on method based on knowledge, ontology and resource model. Conceptual dependency trees are set up based on the ontology theory to denote the relation of concepts. By detailed analysis of the intension and the extension of concepts, pronoun and dynamic pronoun classes templates are established on method based on knowledge. Accomplished the understanding of Chinese pronoun. Forth, Semantic analysis program is written to understand sentence on the basis of the work done, which can understand Chinese pronoun in corresponding fields. Last, Semantic analysis of natural language understanding is applied to requirement analysis archetypal system for the design of cam machine. The pronoun in user requirements is identified. With other models running, the system runs well after debug.
引文
[1] 顾佩华. 设计理论与方法学研究方面的最新进展. 机械与电子,1998.5:26-31
[2] 谢友柏. 现代设计与知识获取. 中国机械工程第 7 卷,第 6 期,1996.
[3] 檀润华.创新设计—TRIZ:发明问题解决理论.北京,机械工业出版社,2002.2
[4] 王知行,刘廷荣. 机械原理. 北京:高等教育出版社,2000
[5] 索东梅.自然语言理解研究.长春师范学院学报,2005.3:110-111
[6] 马健,张珏成.对机器自然语言理解本质的认识.宁夏大学学报,2003.9 第24卷第3期:265-268
[7] Applet,D.E.Studies in Natural Language Processing: Planning English Sentences. Cambrige, U.K.: Cambrige U.Press.1985
[8] 王开铸.自然语言理解――计算机能思维吗?. 哈尔滨:哈尔滨工业大学出版社,1996
[9] 郭艳华,周昌乐.自然语言理解研究综述.杭州电子工业学院学报,2000.2 第20卷第1期:58-64
[10] 王小波.自然语言理解研究.山西广播电视大学学报,2005.4 第2期:27-28
[11] James Allen. Natural Language Understanding. The Benjamin/Cummings Publishing Company, Inc. 1995
[12] 蔡义发.自然语言理解的研究与发展.计算机应用与软件, 1992,第7期:1-6
[13] 赵振西,李凯,金铭源. 汉语的分层理解及其知识表示. 中文信息,1990,第 1 期:16-22
[14] 魏晓宁.人工智能在自然语言理解技术上的应用.科技论坛,中国科技信息2005年第19期
[15] 高文利.自然语言理解的新思路:HNC理论.湖南城市学院学报,2004.11 第25卷第6期:84-86
[16] Pollard,C.,and I.Sag. Information-Based Syntax and Semantics. Vol.I : Fundamentals. CSLI Lecture Notes 13. Chicago: Chicago U. Press. 1987
[17] 贾彦德.汉语语义学.北京大学出版社,1992年11月
[18] D. Gildea and D. Jurafsky. Automatic labeling of semantic roles. Comput. Linguist., 28(3):245-288, 2002.
[19] N. Xue and M. Palmer. Automatic semantic role labeling for Chinese verbs. In Proceedings of IJCAI 2005, 2005.
[20] 李思强,袁满,张连滨等 . 语义对象及建模技术 . 计算机应用研究,1999,第5期: 40-41.
[21] 徐天任,夏幼明,甘健侯等. 用语义网络语言描述知识的表示. 云南师范大学学报,2003.5 第23卷第3期:9-13
[22] 夏幼明. 基于语义网络的知识表示的形式转换及推理. 武汉大学学报(信息科学版) , 2001.26 (4) :369- 372.
[23] 王永庆 . 人工智能原理与方法 . 西安:西安交通大学出版社,1998.
[24] 程琪龙.试论Fillmore格语法1971模式.解放军外国语学院学报,1995,第6期:16-20
[25] 傅承德. 自然语言理解的方法与策略. 郑州:河南人民出版社,1990
[26] OO STHU IZEN G D. Rough sets and concept lattices[A ]. In: Ziarko W Ped. Rough Sets, and Fuzzy Sets and Know ledge Discovery (RSKD’93) [C ]. London: Springer-Verlag, 1994, 24-31.
[27] 黄伟,金远平.形式概念分析在本体构建中的应用.微机发展,2005.2 第15卷第2期:28-31.
[28] 杨秋芬,陈跃新. Ontology方法学综述.计算机应用研究,2002.4
[29] 张宇翔.NKI本体理论中一些基本关系的研究.云南师范大学硕士研究生学位论文,2002.6
[30] William, S. And Austin, T. 1999. Ontologies. IEEE Intelligent Systems, 1999 Jan/Feb: 18-19
[31] M. Uschold and A. Tate, 1998. Putting ontologies to use. The Knowledge Engineering Review, Vol. 13:1: 1-3.
[32] Ontology infrastructure for execution-oriented autonomous agents. RobertE.Wray, SeanA.Lisse, JonathanT.Beard, Robotics and Autonomous Systems, 2004
[33] A rough set model with ontologies for discovering maximal association rules in document collections . YaxinBi*,Terry Anderson, Sally Mc Clean Knowledge-Based Systems, 2003
[34] Stumme G, Maedche A. FCA-MERGE: Bottom-up Merging of Ontologies .In IJCAI, 2001: 225-234
[35] 彭宣维.代词的语篇语法属性、范围及其语义功能分类. 语言教学与研究,2005 年第1 期:56-65
[36] 夏蓉.指代与语域. 西安外国语学院学报, 2004.6 第12卷第2期:12-14
[37] Mitkov R. Anaphora Resolution. London: Longman Press, 2002.
[38] 吕叔湘.近代汉语指代词.学林出版社,1985
[39] 孙蕾.指示性代词的语义特性. 外语学刊, 2002年第3期:100-105
[40] 蒋华,廖艳君,杨安红. 现代汉语中的称代代词与指示代词. 湖南大学学报( 社会科学版),2004.5 第18 卷 第3 期:66-69
[41] 蒋华.现代汉语“这/那”类指示代词的多维度考察.湖南师范大学博士学位论文,2004.3
[42] 曹军,周经野.汉语第三人称代词消解方法研究——一种基于语义结构和中心模型的新方法.湘潭大学硕士学位论文,2002.4
[43] Hobby,J.R.Resolving pronoun references, Lingua 44,B11-338.Reprinted in RNLP.1978
[44] Wang HF. Survey: Computational models and methods of anaphora resolution. Journal of Chinese Information Processing, 2002,16(6):9?17 (in Chinese with English abstract).
[45] 靳焱, 倪兰.疑问代词研究综述.中南民族大学学报(人文社会科版),2003.8 第23卷:273-275
[46] 邹红霆.语义结构和中心理论相结合的代词消解.湖南科技学院学报,2005.5 第26 卷 第5 期:143-146
[47] 吴为章,田小琳《汉语句群》.北京:商务印书馆,2000.(81-82)
[48] 王志.篇章代词“它”用法探析.世界汉语教学,1998年第3期:37-42
[49] Weng,Fu-liang,Wang,Ye-yi .Introduction to Computational Linguistics. The Chinese Academy of Social Press,1998(in Chinese).
[50] 檀润华,王庆禹. 产品设计过程模型、策略与方法综述. 概念设计,2002.11.
[51] KarlCzajkowskiy, etal. Grid Information Services for Distributed Resource Sharing. Proc. 10th IEEE International Symposium on High Performance Distributed Computing (HPDC210) , IEEE Press, 2001.
[52] Xiao Liaobing, Huang Linpeng. Technology of Grid Computing. Computer Engineering, 2002
[53] Reinhart,T. Co-reference and bound anaphora: A restatement of the anaphora question . Linguistics and Philosophy, 1983 6:47-88
[54] Charniak. Towards a model of children’s story comprehension. Massachusetts institute of Technology, 1972
[55] Winograd. Understanding Natural Language Academic Press, 1972
[56] Grosz. Attentions, intentions and the structure of discourse. Computational Linguistics, 1986
[57] Grosz. Towards a computational theory of discourse interpretation. Computational Linguistics, 1986
[58] Kameyama ,M. Stressed and untress pronouns : complementary preferences. In :peter Bosch and Rob van der Sandt (editors) . Focus : linguistic , cognitive , and computational perspectives. Cambridge university Press ,1999 :306 - 321
[2] 谢友柏. 现代设计与知识获取. 中国机械工程第 7 卷,第 6 期,1996.
[3] 檀润华.创新设计—TRIZ:发明问题解决理论.北京,机械工业出版社,2002.2
[4] 王知行,刘廷荣. 机械原理. 北京:高等教育出版社,2000
[5] 索东梅.自然语言理解研究.长春师范学院学报,2005.3:110-111
[6] 马健,张珏成.对机器自然语言理解本质的认识.宁夏大学学报,2003.9 第24卷第3期:265-268
[7] Applet,D.E.Studies in Natural Language Processing: Planning English Sentences. Cambrige, U.K.: Cambrige U.Press.1985
[8] 王开铸.自然语言理解――计算机能思维吗?. 哈尔滨:哈尔滨工业大学出版社,1996
[9] 郭艳华,周昌乐.自然语言理解研究综述.杭州电子工业学院学报,2000.2 第20卷第1期:58-64
[10] 王小波.自然语言理解研究.山西广播电视大学学报,2005.4 第2期:27-28
[11] James Allen. Natural Language Understanding. The Benjamin/Cummings Publishing Company, Inc. 1995
[12] 蔡义发.自然语言理解的研究与发展.计算机应用与软件, 1992,第7期:1-6
[13] 赵振西,李凯,金铭源. 汉语的分层理解及其知识表示. 中文信息,1990,第 1 期:16-22
[14] 魏晓宁.人工智能在自然语言理解技术上的应用.科技论坛,中国科技信息2005年第19期
[15] 高文利.自然语言理解的新思路:HNC理论.湖南城市学院学报,2004.11 第25卷第6期:84-86
[16] Pollard,C.,and I.Sag. Information-Based Syntax and Semantics. Vol.I : Fundamentals. CSLI Lecture Notes 13. Chicago: Chicago U. Press. 1987
[17] 贾彦德.汉语语义学.北京大学出版社,1992年11月
[18] D. Gildea and D. Jurafsky. Automatic labeling of semantic roles. Comput. Linguist., 28(3):245-288, 2002.
[19] N. Xue and M. Palmer. Automatic semantic role labeling for Chinese verbs. In Proceedings of IJCAI 2005, 2005.
[20] 李思强,袁满,张连滨等 . 语义对象及建模技术 . 计算机应用研究,1999,第5期: 40-41.
[21] 徐天任,夏幼明,甘健侯等. 用语义网络语言描述知识的表示. 云南师范大学学报,2003.5 第23卷第3期:9-13
[22] 夏幼明. 基于语义网络的知识表示的形式转换及推理. 武汉大学学报(信息科学版) , 2001.26 (4) :369- 372.
[23] 王永庆 . 人工智能原理与方法 . 西安:西安交通大学出版社,1998.
[24] 程琪龙.试论Fillmore格语法1971模式.解放军外国语学院学报,1995,第6期:16-20
[25] 傅承德. 自然语言理解的方法与策略. 郑州:河南人民出版社,1990
[26] OO STHU IZEN G D. Rough sets and concept lattices[A ]. In: Ziarko W Ped. Rough Sets, and Fuzzy Sets and Know ledge Discovery (RSKD’93) [C ]. London: Springer-Verlag, 1994, 24-31.
[27] 黄伟,金远平.形式概念分析在本体构建中的应用.微机发展,2005.2 第15卷第2期:28-31.
[28] 杨秋芬,陈跃新. Ontology方法学综述.计算机应用研究,2002.4
[29] 张宇翔.NKI本体理论中一些基本关系的研究.云南师范大学硕士研究生学位论文,2002.6
[30] William, S. And Austin, T. 1999. Ontologies. IEEE Intelligent Systems, 1999 Jan/Feb: 18-19
[31] M. Uschold and A. Tate, 1998. Putting ontologies to use. The Knowledge Engineering Review, Vol. 13:1: 1-3.
[32] Ontology infrastructure for execution-oriented autonomous agents. RobertE.Wray, SeanA.Lisse, JonathanT.Beard, Robotics and Autonomous Systems, 2004
[33] A rough set model with ontologies for discovering maximal association rules in document collections . YaxinBi*,Terry Anderson, Sally Mc Clean Knowledge-Based Systems, 2003
[34] Stumme G, Maedche A. FCA-MERGE: Bottom-up Merging of Ontologies .In IJCAI, 2001: 225-234
[35] 彭宣维.代词的语篇语法属性、范围及其语义功能分类. 语言教学与研究,2005 年第1 期:56-65
[36] 夏蓉.指代与语域. 西安外国语学院学报, 2004.6 第12卷第2期:12-14
[37] Mitkov R. Anaphora Resolution. London: Longman Press, 2002.
[38] 吕叔湘.近代汉语指代词.学林出版社,1985
[39] 孙蕾.指示性代词的语义特性. 外语学刊, 2002年第3期:100-105
[40] 蒋华,廖艳君,杨安红. 现代汉语中的称代代词与指示代词. 湖南大学学报( 社会科学版),2004.5 第18 卷 第3 期:66-69
[41] 蒋华.现代汉语“这/那”类指示代词的多维度考察.湖南师范大学博士学位论文,2004.3
[42] 曹军,周经野.汉语第三人称代词消解方法研究——一种基于语义结构和中心模型的新方法.湘潭大学硕士学位论文,2002.4
[43] Hobby,J.R.Resolving pronoun references, Lingua 44,B11-338.Reprinted in RNLP.1978
[44] Wang HF. Survey: Computational models and methods of anaphora resolution. Journal of Chinese Information Processing, 2002,16(6):9?17 (in Chinese with English abstract).
[45] 靳焱, 倪兰.疑问代词研究综述.中南民族大学学报(人文社会科版),2003.8 第23卷:273-275
[46] 邹红霆.语义结构和中心理论相结合的代词消解.湖南科技学院学报,2005.5 第26 卷 第5 期:143-146
[47] 吴为章,田小琳《汉语句群》.北京:商务印书馆,2000.(81-82)
[48] 王志.篇章代词“它”用法探析.世界汉语教学,1998年第3期:37-42
[49] Weng,Fu-liang,Wang,Ye-yi .Introduction to Computational Linguistics. The Chinese Academy of Social Press,1998(in Chinese).
[50] 檀润华,王庆禹. 产品设计过程模型、策略与方法综述. 概念设计,2002.11.
[51] KarlCzajkowskiy, etal. Grid Information Services for Distributed Resource Sharing. Proc. 10th IEEE International Symposium on High Performance Distributed Computing (HPDC210) , IEEE Press, 2001.
[52] Xiao Liaobing, Huang Linpeng. Technology of Grid Computing. Computer Engineering, 2002
[53] Reinhart,T. Co-reference and bound anaphora: A restatement of the anaphora question . Linguistics and Philosophy, 1983 6:47-88
[54] Charniak. Towards a model of children’s story comprehension. Massachusetts institute of Technology, 1972
[55] Winograd. Understanding Natural Language Academic Press, 1972
[56] Grosz. Attentions, intentions and the structure of discourse. Computational Linguistics, 1986
[57] Grosz. Towards a computational theory of discourse interpretation. Computational Linguistics, 1986
[58] Kameyama ,M. Stressed and untress pronouns : complementary preferences. In :peter Bosch and Rob van der Sandt (editors) . Focus : linguistic , cognitive , and computational perspectives. Cambridge university Press ,1999 :306 - 321