基于概念网的智能信息服务
|
摘要
随着人类社会步入信息社会，人类面临“信息爆炸”、“混沌信息空间”和“数据过剩”带来的巨大压力。人们迫切需要一种智能代理完成多种耗时、复杂的工作。针对上述问题，本课题旨在设计一个基于概念网络的智能信息服务器，帮助用户有效的收集、检索和管理信息，为用户提供高效的信息服务。
本课题采用概念网络作为其后台知识库。概念网络是一个描述客观事物及其内在联系的一个信息模型。该模型以一个对象的观点来认识概念的内涵，把概念看成是一个具有属性、行为、概念描述、概念词的复杂对象体。概念与概念之间再通过关联关系构成一个复杂的概念网络。通过这个网络，知识之间便构建了一种“联想的脉络，推理的依据”。
本文详细地讨论智能信息服务器的整体构建策略。系统依靠后台的概念网络，处理信息的收集和检索请求。对用户收集的文档采用概念分词算法、概念特征和角色提取算法，将文档进行自动分类和关联，从而有效、合理的组织文档。在接收到用户的查询请求时，系统通过对用户查询意图的理解，过滤与用户查询意图不相关的文档，从而有效提高查询精度。为了提高检索的精度和效率，系统采用交互式的检索模式，逐步聚焦检索目标。
本文还针对本题中概念网络匹配、个性化服务和Web服务三个关键技术做了重点讨论。给出了一套概念网的概念运算方法，通过运用这些概念运算方法完成各种概念匹配任务。具体讨论了用户个性模式的创建、表示和学习算法，辅助本系统支持个性化服务。对当前热门的Web服务技术从它的整体服务体系结构开始，到它的开发周期以及应用程序和服务、服务与服务间的消息传递机制做了详细的阐述。
最后，本文分别从用户接口层、业务对象层和数据层介绍系统的具体实现。给出了实验结果证明本文的观点，并对今后的工作提出新的要求。
When human society has come into an information age, people faced great stress brought by "Information Explosion", "Information Chaotic Space" and "Data Glut". People need an intelligent agent to assist them to complete all kinds of time-consuming and complex tasks. Our topic is aimed to design an intelligent information server based on conceptual network, which possesses information collection, retrieval and management functions, and provides high effective information services.
To achieve intelligent information service, a powerful repository is necessary. In our paper, we selected conceptual network as the domain model to construct knowledge. Conceptual network is used to describe the objective things and their relations. This model describes concept as a complex object with attribute, behavior, description, and concept synonymies. We construct a conceptual network by describing the structural and semantic relationships among concepts. Using this network we get " a grain of association of thought and a reliance of reasoning " among all the knowledge units.
Our paper discusses construction strategy about intelligent information server in detail. The system processes information collecting and retrieving requests relying on conceptual network. We use word segmentation algorithm, conceptual features and role extraction algorithm to accomplish document auto-classification and co-relation, and then consequently attain to effective and reasonable organizing documents. When system received users' retrieving request, it could understand user's intention, filtrate irrelevant documents, and then fulfill user's needs. In order to enhance retrieval's precision and efficiency, system adopted mutual search patterns to precise field of retrieval step by step.
The paper also emphasizes on discussing three key techniques: concept matching, personalized service and Web services. We give a set
of concept operations, and we use them to accomplish various concept-matching tasks. Concretely, we use user's profile to enable our system to support personalized services. Also we discuss its construction, representation and learning algorithm. Nowadays Web services is fashionable development technique. We explain its whole architecture, development lifecycle and messaging mechanism.
At last, this paper introduces the system's implementation with user interface layer, business layer and data layer separately. It gives the experiment result to testify our paper's viewpoints, and also brings forward the new research work in future.
本课题采用概念网络作为其后台知识库。概念网络是一个描述客观事物及其内在联系的一个信息模型。该模型以一个对象的观点来认识概念的内涵，把概念看成是一个具有属性、行为、概念描述、概念词的复杂对象体。概念与概念之间再通过关联关系构成一个复杂的概念网络。通过这个网络，知识之间便构建了一种“联想的脉络，推理的依据”。
本文详细地讨论智能信息服务器的整体构建策略。系统依靠后台的概念网络，处理信息的收集和检索请求。对用户收集的文档采用概念分词算法、概念特征和角色提取算法，将文档进行自动分类和关联，从而有效、合理的组织文档。在接收到用户的查询请求时，系统通过对用户查询意图的理解，过滤与用户查询意图不相关的文档，从而有效提高查询精度。为了提高检索的精度和效率，系统采用交互式的检索模式，逐步聚焦检索目标。
本文还针对本题中概念网络匹配、个性化服务和Web服务三个关键技术做了重点讨论。给出了一套概念网的概念运算方法，通过运用这些概念运算方法完成各种概念匹配任务。具体讨论了用户个性模式的创建、表示和学习算法，辅助本系统支持个性化服务。对当前热门的Web服务技术从它的整体服务体系结构开始，到它的开发周期以及应用程序和服务、服务与服务间的消息传递机制做了详细的阐述。
最后，本文分别从用户接口层、业务对象层和数据层介绍系统的具体实现。给出了实验结果证明本文的观点，并对今后的工作提出新的要求。
When human society has come into an information age, people faced great stress brought by "Information Explosion", "Information Chaotic Space" and "Data Glut". People need an intelligent agent to assist them to complete all kinds of time-consuming and complex tasks. Our topic is aimed to design an intelligent information server based on conceptual network, which possesses information collection, retrieval and management functions, and provides high effective information services.
To achieve intelligent information service, a powerful repository is necessary. In our paper, we selected conceptual network as the domain model to construct knowledge. Conceptual network is used to describe the objective things and their relations. This model describes concept as a complex object with attribute, behavior, description, and concept synonymies. We construct a conceptual network by describing the structural and semantic relationships among concepts. Using this network we get " a grain of association of thought and a reliance of reasoning " among all the knowledge units.
Our paper discusses construction strategy about intelligent information server in detail. The system processes information collecting and retrieving requests relying on conceptual network. We use word segmentation algorithm, conceptual features and role extraction algorithm to accomplish document auto-classification and co-relation, and then consequently attain to effective and reasonable organizing documents. When system received users' retrieving request, it could understand user's intention, filtrate irrelevant documents, and then fulfill user's needs. In order to enhance retrieval's precision and efficiency, system adopted mutual search patterns to precise field of retrieval step by step.
The paper also emphasizes on discussing three key techniques: concept matching, personalized service and Web services. We give a set
of concept operations, and we use them to accomplish various concept-matching tasks. Concretely, we use user's profile to enable our system to support personalized services. Also we discuss its construction, representation and learning algorithm. Nowadays Web services is fashionable development technique. We explain its whole architecture, development lifecycle and messaging mechanism.
At last, this paper introduces the system's implementation with user interface layer, business layer and data layer separately. It gives the experiment result to testify our paper's viewpoints, and also brings forward the new research work in future.
引文
[1]G. W. Fumas, T. K. Landauer, L. M. Gomez, et al. The Vocabulary Problem in Human-System Communication. Communications of the ACM,1987, 30(11): 964～971
[2]Apte C, Damerau J F, Weiss S. Automated learning of decision rules for text categorization. ACM Transactions on Information System, 1994, 12(3): 233～251
[3]Yang Y. Expert network: Effective and efficient learning from human decisions in text categorization and retrieval. In: Proc of the Seventeenth Int'1 ACM SIGIR Conf on Research and Development in Information Retrieval. Dublin, 1994. 13～22
[4]Lewis D D, Schapore R E,Callan J P, et al. Training algorithms for linear text classifiers. In: Proceeding of the Nineteenth Int'l ACM SIGIR Conf on Research and Development in Information Retrieval. Zurich, 1996. 298～306
[5]Cohen W W, Singer Y. Context-sensitive learning methods for text categorization. In: Proceeding of the 19th Int'l ACM SIGIR Conf on Research and Development in Information Retrieval. Zurich,1996. 307～315
[6]Lewis D D. Naive (Bayes) at forty: The independence assumption in information retrieval. In Machine Learning: Tenth European Conference on Machine Learning(ECML-98). Chemnitz,DE, 1998.4～15
[7]赖茂生等．计算机情报检索．北京：北京大学出版社，1993
[8]罗三定，黄勇．一个应用模糊方法的智能搜索引擎的构建．计算机工程，2000，26(12)：113～115
[9]罗三定，黄勇．一个基于具有自学习机制的概念网络的搜索引擎．计算机工程，2001，27(9)：89～92
[10]陆文彦．概念网的建模、实现与应用：[硕士学位论文]．湖南长沙：中南大学，2002
[11]罗三定，冯元勇，沈德耀 等．基于概念的文档评价模型．计算机工程，2002，28(8)：78～80，283
[12]冯元勇．职能搜索器的概念库设计及其文档评价策略：[硕士学位论文]．湖南长沙：中南大学，2002
[13]Sanding Luo, Sha Sha, Deyao Shen, et al. Conceptual Network Based
Courseware Navigation and Web Presentation Mechanisms. In: Springer 2436 Lecture Notes in Computer Science,2002.81～94
[14]罗三定，陆文彦，王浩等．基于概念的文本类别特征提取与文本模糊匹配．计算机工程与应用，2002，38(16)：97～99
[15]沙莎，王浩，陆文彦等．基于作者关联和概念网的科技论文搜索方法研究．计算机工程，2002，28(12)：126～128
[16]董振东，董强．知网．http://www.keenage.com/zhiwang/c_zhiwang.html
[17]黄曾阳．HNC(概念层次网络)理论—计算机理解语言研究的新思路(http://www.hncnlp.com/)．北京：清华大学出版社，1998．1～110
[18]J. F. Sowa. Conceptual Structure: Information Processing in Mind and Machine Reading. MA: Addsion-Wesly Publishing Co, 1984
[19]J. F. Sowa. Conceptual graphs for a database inference. IBM J.Res. Develop, 1986,30(1): 57～69
[20]P. H. de Vries. Representation of Science Texts in Knowledge Graphs: [Ph.D.'s thesis].Enschede, The Netherlands: University ofTwente, 1989
[21]陆汝钤，石纯一，张松懋等．面向Agent的常识知识库．中国科学(E辑)，2000，30(5)：453～463
[22]N. J. Belkin, R. N. Oddy, H. M. Brooks. Ask for information retrieval: Part I background and theory. Joumal of Documentation, 1982, 38(2): 61～71
[23]H. Chen, T. D. Ng, J. Martinez, et al. A Concept Space Approach to Addressing the Vocabulary Problem in Scientific Information Retrieval: An Experiment on the Worm Community System. Journal of the American Society for Information Science, 1997,48(1): 17～31
[24]Shian-Hua Lin, Meng Chang Chen, Jan-Ming Ho, et al. ACIRD: Intelligent Intemet Documents Organization and Retrieval. IEEE Transactions on Knowledge and Data Engineering,2002,14(3): 599～614
[25]蔡自兴，徐光佑．人工智能及其应用．北京：清华大学出版社，1996．17～50
[26]李晓黎，刘继敏，史忠植．概念推理网及其在文本分类中的应用．计算机研究与发展，2000，37(9)：1032～1038
[27]Eui-Hong Han G, Karypis. Concept indexing-a fast dimensionality reducing algorithm with applications to document retrieval and categorization. Technical report, University of Minnesota, Dept. of Computer Science / Army HPC Research Center,
March 2000
[28]陆玉昌，鲁明羽，李凡等．向量空间法中单词权重函数的分析和构造．计算机研究与发展，2002，39(10)：1205～1210
[29]庞剑锋，卜东波，白硕．基于向量空间模型的文本自动分类系统的研究与实现．计算机应用研究，2001，(9)：23～26
[30]Guus Schreiber等．知识工程和知识管理(史忠植，梁永全，吴斌等译)．北京：机械工业出版社，2003
[31]危辉，潘云鹤．从知识表示到表示：人工智能认识论上的进步．计算机研究与发展，2000，37(7)：819～825
[32]武成岗，焦文品，田启家等．基于本体论和多主体的信息检索服务器．计算机研究与发展，2001，38(6)：641～647
[33]郑毅，吴斌，史忠植．基于概念空间的文本检索系统．计算机工程与应用，2002，(12)：67～69
[34]艾丹祥，张玉峰．相关反馈技术在知识检索中的应用．情报科学，2003，21(10)：1100～1103
[35]A Pretschner. Ontology Based Personalized Search: [Master's thesis]. USA: The University of Kansas, 1999
[36]Ashish Banerjee，Aravind Corera等．C≠Web服务高级编程(康博译)．北京：清华大学出版社，2002
[37]Etban Cerami. Web服务精髓(陈逸译)．北京：中国电力出版社，2003
[38]王钰，袁小红，石纯一等．关于知识表示的讨论．计算机学报，1995，18(3)：212～224
[39]傅伟鹏，吴斌，何清等．一种概念空间自生成方法．计算机工程与应用，2002，(7)：63～65，88
[40]赵亮，胡乃静，张守志．个性化推荐算法设计．计算机研究与发展，2002，39(8)：986～991
[41]林鸿飞，杨元生．用户兴趣模型的表示和更新机制．计算机研究与发展，2002，39(7)：843～847
[42]Haverkamp D, Gauch S. Intelligent Infbrmation Agents: Review and Challenges for Distributed Information Sources. Journal of the American Society for Information Science, 1998, 49(4):304～311
[43]Bagga, Amit, and Breck Baldwin. Entity-Based Cross-Document
Coreferencing Using the Vector Space Model. In: Proceedings of the 36th Annual Meeting of the Association for Computational Linguistics and the 17th International Conference on Computational Linguistics (COLINGACL '98). 1998.79～85
[44]李源，何清，史忠植．基于概念语义空间的联想检索．北京科技大学学报，2001，23(6)：577～580
[45]林鸿飞．基于混合模式的文本过滤模型．计算机研究与发展，2001，38(9)：1127～1131
[46]王继成，萧嵘，孙正兴等．Web信息检索研究进展．计算机研究与发展，2001，38(2)：187～193
[47]Jiawei Han, Mieheline Kamber. Data Mining: Concepts and Techniques.北京：高等教育出版社，2001
[2]Apte C, Damerau J F, Weiss S. Automated learning of decision rules for text categorization. ACM Transactions on Information System, 1994, 12(3): 233～251
[3]Yang Y. Expert network: Effective and efficient learning from human decisions in text categorization and retrieval. In: Proc of the Seventeenth Int'1 ACM SIGIR Conf on Research and Development in Information Retrieval. Dublin, 1994. 13～22
[4]Lewis D D, Schapore R E,Callan J P, et al. Training algorithms for linear text classifiers. In: Proceeding of the Nineteenth Int'l ACM SIGIR Conf on Research and Development in Information Retrieval. Zurich, 1996. 298～306
[5]Cohen W W, Singer Y. Context-sensitive learning methods for text categorization. In: Proceeding of the 19th Int'l ACM SIGIR Conf on Research and Development in Information Retrieval. Zurich,1996. 307～315
[6]Lewis D D. Naive (Bayes) at forty: The independence assumption in information retrieval. In Machine Learning: Tenth European Conference on Machine Learning(ECML-98). Chemnitz,DE, 1998.4～15
[7]赖茂生等．计算机情报检索．北京：北京大学出版社，1993
[8]罗三定，黄勇．一个应用模糊方法的智能搜索引擎的构建．计算机工程，2000，26(12)：113～115
[9]罗三定，黄勇．一个基于具有自学习机制的概念网络的搜索引擎．计算机工程，2001，27(9)：89～92
[10]陆文彦．概念网的建模、实现与应用：[硕士学位论文]．湖南长沙：中南大学，2002
[11]罗三定，冯元勇，沈德耀 等．基于概念的文档评价模型．计算机工程，2002，28(8)：78～80，283
[12]冯元勇．职能搜索器的概念库设计及其文档评价策略：[硕士学位论文]．湖南长沙：中南大学，2002
[13]Sanding Luo, Sha Sha, Deyao Shen, et al. Conceptual Network Based
Courseware Navigation and Web Presentation Mechanisms. In: Springer 2436 Lecture Notes in Computer Science,2002.81～94
[14]罗三定，陆文彦，王浩等．基于概念的文本类别特征提取与文本模糊匹配．计算机工程与应用，2002，38(16)：97～99
[15]沙莎，王浩，陆文彦等．基于作者关联和概念网的科技论文搜索方法研究．计算机工程，2002，28(12)：126～128
[16]董振东，董强．知网．http://www.keenage.com/zhiwang/c_zhiwang.html
[17]黄曾阳．HNC(概念层次网络)理论—计算机理解语言研究的新思路(http://www.hncnlp.com/)．北京：清华大学出版社，1998．1～110
[18]J. F. Sowa. Conceptual Structure: Information Processing in Mind and Machine Reading. MA: Addsion-Wesly Publishing Co, 1984
[19]J. F. Sowa. Conceptual graphs for a database inference. IBM J.Res. Develop, 1986,30(1): 57～69
[20]P. H. de Vries. Representation of Science Texts in Knowledge Graphs: [Ph.D.'s thesis].Enschede, The Netherlands: University ofTwente, 1989
[21]陆汝钤，石纯一，张松懋等．面向Agent的常识知识库．中国科学(E辑)，2000，30(5)：453～463
[22]N. J. Belkin, R. N. Oddy, H. M. Brooks. Ask for information retrieval: Part I background and theory. Joumal of Documentation, 1982, 38(2): 61～71
[23]H. Chen, T. D. Ng, J. Martinez, et al. A Concept Space Approach to Addressing the Vocabulary Problem in Scientific Information Retrieval: An Experiment on the Worm Community System. Journal of the American Society for Information Science, 1997,48(1): 17～31
[24]Shian-Hua Lin, Meng Chang Chen, Jan-Ming Ho, et al. ACIRD: Intelligent Intemet Documents Organization and Retrieval. IEEE Transactions on Knowledge and Data Engineering,2002,14(3): 599～614
[25]蔡自兴，徐光佑．人工智能及其应用．北京：清华大学出版社，1996．17～50
[26]李晓黎，刘继敏，史忠植．概念推理网及其在文本分类中的应用．计算机研究与发展，2000，37(9)：1032～1038
[27]Eui-Hong Han G, Karypis. Concept indexing-a fast dimensionality reducing algorithm with applications to document retrieval and categorization. Technical report, University of Minnesota, Dept. of Computer Science / Army HPC Research Center,
March 2000
[28]陆玉昌，鲁明羽，李凡等．向量空间法中单词权重函数的分析和构造．计算机研究与发展，2002，39(10)：1205～1210
[29]庞剑锋，卜东波，白硕．基于向量空间模型的文本自动分类系统的研究与实现．计算机应用研究，2001，(9)：23～26
[30]Guus Schreiber等．知识工程和知识管理(史忠植，梁永全，吴斌等译)．北京：机械工业出版社，2003
[31]危辉，潘云鹤．从知识表示到表示：人工智能认识论上的进步．计算机研究与发展，2000，37(7)：819～825
[32]武成岗，焦文品，田启家等．基于本体论和多主体的信息检索服务器．计算机研究与发展，2001，38(6)：641～647
[33]郑毅，吴斌，史忠植．基于概念空间的文本检索系统．计算机工程与应用，2002，(12)：67～69
[34]艾丹祥，张玉峰．相关反馈技术在知识检索中的应用．情报科学，2003，21(10)：1100～1103
[35]A Pretschner. Ontology Based Personalized Search: [Master's thesis]. USA: The University of Kansas, 1999
[36]Ashish Banerjee，Aravind Corera等．C≠Web服务高级编程(康博译)．北京：清华大学出版社，2002
[37]Etban Cerami. Web服务精髓(陈逸译)．北京：中国电力出版社，2003
[38]王钰，袁小红，石纯一等．关于知识表示的讨论．计算机学报，1995，18(3)：212～224
[39]傅伟鹏，吴斌，何清等．一种概念空间自生成方法．计算机工程与应用，2002，(7)：63～65，88
[40]赵亮，胡乃静，张守志．个性化推荐算法设计．计算机研究与发展，2002，39(8)：986～991
[41]林鸿飞，杨元生．用户兴趣模型的表示和更新机制．计算机研究与发展，2002，39(7)：843～847
[42]Haverkamp D, Gauch S. Intelligent Infbrmation Agents: Review and Challenges for Distributed Information Sources. Journal of the American Society for Information Science, 1998, 49(4):304～311
[43]Bagga, Amit, and Breck Baldwin. Entity-Based Cross-Document
Coreferencing Using the Vector Space Model. In: Proceedings of the 36th Annual Meeting of the Association for Computational Linguistics and the 17th International Conference on Computational Linguistics (COLINGACL '98). 1998.79～85
[44]李源，何清，史忠植．基于概念语义空间的联想检索．北京科技大学学报，2001，23(6)：577～580
[45]林鸿飞．基于混合模式的文本过滤模型．计算机研究与发展，2001，38(9)：1127～1131
[46]王继成，萧嵘，孙正兴等．Web信息检索研究进展．计算机研究与发展，2001，38(2)：187～193
[47]Jiawei Han, Mieheline Kamber. Data Mining: Concepts and Techniques.北京：高等教育出版社，2001