摘要
21世纪是海洋的世纪。人类生活离不开海洋,海洋为人类提供丰富的食物资源和矿产资源,是人类生存和发展的基础。海洋是容纳热量的大热库,海洋碳循环效率对全球气候的变化有着深远的影响。人类很早就开始了对海洋的探索,并在海洋探索中积累知识,收集数据和资料。自1872-1876年英国“挑战者”号环球海洋考察开创真正意义上的海洋学研究以来,海洋科学研究机构越来越多,观测设备越来越先进,观测手段越来越丰富。至今,各种海洋研究机构存储了大量的海洋数据和资料,这些数据和资料对进一步的海洋探索有着重要的参考价值和理论指导意义。
然而,各种研究机构的数据存储格式并不一致,对数据的共享起到了阻碍作用。海洋元数据可以较好地解决这个问题,能够实现海洋数据的互操作。
但是,各研究机构的数据和资料还存在一词多义、同义词、术语分类不明确等问题,元数据并不能很好地理解和处理这些语义。因此,必须寻找一种新的途径来实现语义的理解。
本体作为共享概念模型的明确的形式化规范说明,可以捕获相关领域的知识,提供对该领域知识的共同理解,确定该领域内共同认可的词汇,并从不同层次的形式化模式给出这些词汇或术语以及词汇间相互关系的明确定义。因此,本体可以担此重任。
本体以其语义优势在很多领域都有着广泛的应用,尤其是在信息检索方面,可以提高检索的查全率和查准率,因此,本文建立了一个海洋生态本体,并在此本体的基础上开发了一个优化的信息检索系统,实验证明,此系统可以有效提高涉海信息检索的效率。
海洋生态是一个复杂的系统,海洋中的各生物因子和非生物因子以及生物因子之间相互依赖、相互作用、相互影响。同时,海洋中充满了角色,如捕食者与被捕食者、生产者与消费者等等,甚至,一个实体可以扮演多个角色,如浮游植物在光合作用中充当化学反应的主体,在细菌分解作用中又扮演着分解对象的角色；氮、磷等营养物质可以加速浮游植物的生长繁殖,起着积极的促进作用,但是,过量的营养物质则会引起赤潮或藻华等海洋灾害,此时的氮、磷等元素又扮演了不利因素的角色。
Hozo的role理论恰好可以描述这些复杂的关系。日本大阪大学沟口实验室在前人的研究基础上提出了自己的role理论,并在这个理论的基础上,开发了一个功能强大的可视化本体编辑工具Hozo。本文的研究工作正是在role理论的基础上开展的,并选择Hozo作为本体开发工具。
本文的研究工作有两大部分：第一,充分利用role理论,建立海洋生态系统的本体模型,并对其中的核心概念进行了形式化建模；第二,基于海洋生态本体开发了一个优化的信息检索系统,该系统能够较好地改进传统搜索引擎在海洋生态领域的检索结果,提高检索效率。
本文的主要内容和组织结构如下：
第一章主要叙述了本课题的研究背景、研究意义和应用价值,并综述了国内外的研究现状。
第二章介绍了本体的定义、分类、描述语言、构建方法和开发工具等方面的理论知识,并对本体的发展动态进行了叙述。
第三章详细介绍了Hozo的本体建模理论,包括Hozo的体系结构、role理论及相关的重要概念,并用丰富的实例阐述了role理论的表达及应用方式。
第四章介绍了海洋生态系统的基本知识,然后借助Hozo开发工具及role理论描述了海洋中各生态因子对海洋生态环境的作用和影响,并对海洋生态环境中的主要现象及作用进行概念建模。
第五章基于本体工程的知识提出了海洋生态本体的构建模型,并利用Hozo完成海洋生态本体的构建。
第六章介绍了OASIS系统的体系结构及采用的优化查询策略,并在JPPF框架下,实现了OASIS系统的分布式并行检索功能。在海洋生态本体的基础上,利用OASIS系统进行了相关实验。实验结果表明,基于海洋生态本体的信息检索系统可以切实提高海洋生态领域信息检索的效率。
第七章对本文的工作内容进行了总结,并指出存在的问题和不足之处。最后对下一步的研究工作进行了展望。
The 21st century is the century of ocean. Ocean provides abundant food and mineral resources for human beings. It is the basis for human survival and development. Ocean is a large thermal reservoir to accommodate the heat. The efficiency of ocean carbon cycle to global climate change has far-reaching impact. Humans began to explore the seas and oceans very early and gained the accumulation of knowledge, data and information. Since the British "Challenger" International Exploration of the Sea between 1872 and 1876 to create a true sense of the oceanographic research, there have been more and more of marine scientific research institutions, more and more advanced observation equipments, and more and more observation methods. So far, a variety of marine research institutions store a large number of marine data and information, which is important to guide further exploration about ocean.
However, the data storage formats of these research institutions are not always consistent. That hinders the sharing of data. Marine metadata can solve this problem, by achieving inter-operability of marine data.
But some of the research data and information has polysemy, synonyms, unclear terminology and so on. Metadata is not well to deal with these semantics. Therefore, we must find a new way to realize the semantic understanding.
Ontology as a formal, explicit specification of a shared conceptualization captures the knowledge of related fields, and provides common understanding in the domain. It can represent concepts and the relationships among them. Semantic within ontology has advantages in many fields and has a wide range of applications, especially in the information retrieval. In these applications, usually a lightweight ontology is sufficient.
However, the marine ecosystem is a very complex system, and the current ontology building method can hardly describe the semantic relationships of various biological factors and non-biological factors and the interaction among them. Meanwhile, the ocean is full of roles, such as predator and prey, producers and consumers. In addition, some entities can play multiple roles. For example, phytoplankton in photosynthesis acts as agent of main chemical reactions, but in the bacterial decomposition plays a role of the object of decomposition. So, to build a more comprising ontology with deep semantics, it is necessary to deal with roles.
Role theory of Hozo can exactly describe these complex relationships. Mizoguchi Lab of Osaka University put forward the role theory referencing other researches, and base on the role theory, they developed a powerful visual ontology editing tool named Hozo. This research is also based on the role theory, and Hozo is used to build marine ecology ontology.
This research work has two parts. Firstly, by using the role theory, we establish a model of marine ecology ontology, and model the core concepts of this ontology with Hozo. Secondly, based on this ontology, an optimized information retrieval system named OASIS is developed. Experiments show that the system improves the retrieval results of traditional search engines.
The main content and organization of this paper are as follows:
Chapter 1 describes the research background, research significance, and historical overview of the application of ontology in marine field.
Chapter 2 describes the definition, classification, description language, construction methods and tools of ontology. The most recently researches on ontology are also described.
Chapter 3 introduces the theory of role modeling which constitutes the basis of Hozo, an ontology editor. The architecture of Hozo and some related concepts are also presented.
Chapter 4 introduces the basic knowledge of marine ecosystems, and then describes the marine ecological factors and their impacts in the marine environment by using Hozo development tools and role theory.
In chapter 5, a marine ecosystem construction model based on ontology engineering is proposed subsequently. The process of the construction of the marine ecosystem ontology and some techniques used in the construction are also presented.
Chapter 6 describes the structure and the optimization strategy of the information retrieval system OASIS. By using the ontology constructed in chapter 5, some experiments are conducted upon traditional search engine and OASIS. Experiments show that the OASIS system improves the precision of the search result in marine field. By using a parallel processing framework named JPPF, we make the OASIS system do retrieval tasks in a distributed parallel way. The acceleration of ontology parsing and searching is analyzed.
Chapter 7 concludes this thesis by summarizing the results, problems and deficiencies about this paper. Some future works are also discussed.
引文
[1]傅桦,吴雁华,曲利娟.生态学原理与应用.1999,
[2]陈泮勤,黄耀,于贵瑞.地球系统碳循环.科学出版社,2004.
[3]冯项云,肖珑,廖三三,庄纪林.国外常用元数据标准比较研究.大学图书馆学报,2001,19(004)：15～21
[4]薛惠芬.国际上几种海洋元数据内容剖析.海洋信息,2004,(003)：25～28
[5]Fonsecaa F, Martinb J. Space and time in eco-ontologies. AI Communications,2004,17(4): 259～269
[6]Fonseca F, Martin J, Rodriguez M. From geo-to eco-ontologies. Geographic Information Science, 2002,:93～107
[7]Tempero E, Biddle R. Simulating multiple inheritance in Java. Journal of Systems and Software, 2000,55(1):87～100
[8]Sunagawa E, Kozaki K, Kitamura Y, et al. Organizing role-concepts in ontology development environment:Hozo. Citeseer,2005
[9]Masolo C, Vieu L, Bottazzi E, et al. Social roles and their descriptions. Procs. of KR'04,2004,: 267～277
[10]Steimann F. On the representation of roles in object-oriented and conceptual modelling. Data & Knowledge Engineering,2000,35(1):83～106
[11]Loebe F. Abstract vs. social roles-Towards a general theoretical account of roles. Applied Ontology,2007,2(2):127～158
[12]Guarino N, Poli R. Formal ontology, conceptual analysis and knowledge representation. International Journal of Human Computer Studies,1995,43(5):625～640
[13]Guarino N. Concepts, attributes and arbitrary relations:Some linguistic and ontological criteria for structuring knowledge bases. Data & Knowledge Engineering,1992,8(3):249～261
[14]Guarino N. Some ontological principles for designing upper level lexical resources. Citeseer, 1998
[15]Kozaki K, Kitamura Y, Ikeda M, et al. Hozo:an environment for building/using ontologies based on a fundamental consideration of "role" and "relationship". Knowledge Engineering and Knowledge Management:Ontologies and the Semantic Web,2002,:155～163
[16]Kozaki K, Kitamura Y, Ikeda M, et al. Development of an Environment for Building Ontologies which is based on a Fundamental Consideration of Relationship" and" Role. Citeseer,2000
[17]Inaba A, Mizoguchi R. Learners'roles and predictable educational benefits in collaborative learning an ontological approach to support design and analysis of CSCL. Springer,2004
[18]Mizoguchi R, Kou H, Zhou J, et al. An Advanced Clinical Ontology.2009,
[19]Kou H, Zhou J, Ohta M, et al. Dynamic is-a Hierarchy Generation from a Clinical Medical Ontology.2009,
[20]Kumazawa T, Saito O, Kozaki K, et al. Toward knowledge structuring of sustainability science based on ontology engineering. Sustainability Science,2009,4(1):99～116
[21]Kumazawa T, Saito O, Kozaki K, et al. Knowledge Structuring Tool for Sustainability Science Based on Ontology Engineering:2009
[22]Fonseca F T, Egenhofer M J. Ontology-Driven Geographic Information Systems. Assn for Computing Machinery,1999
[23]孙敏,陈秀万,张飞舟.地理信息本体论.地理与地理信息科学,2004,20(003)：6～11
[24]杜云艳,周成虎,苏奋振.海岸带及近海科学数据集成与共享研究.2005,
[25]Raskin R. Guide to SWEET ontologies. NASA/Jet Propulsion Lab, Pasadena, CA,2006,
[26]Bermudez L, Graybeal J, Arko R. A marine platforms ontology:Experiences and lessons. Citeseer,2006
[27]Neches R, Fikes R E, Finin T, et al. Enabling technology for knowledge sharing. AI magazine, 1991,12(3):36
[28]Gruber T R. A translation approach to portable ontology specifications. Knowledge acquisition, 1993,5:199
[29]Borst W N. Construction of engineering ontologies for knowledge sharing and reuse.1997,
[30]Studer R, Benjamins V R, Fensel D. Knowledge engineering:principles and methods. Data & Knowledge Engineering,1998,25(1-2):161～197
[31]杨海艳.本体的相似性关系的研究.2009,
[32]Guarino N. Semantic matching:Formal ontological distinctions for information organization, extraction, and integration. Information Extraction A Multidisciplinary Approach to an Emerging Information Technology,1997,:139～170
[33]邓志鸿,唐世渭,张铭,等Ontology研究综述.北京大学学报：自然科学版,2002,38(005)：730～738
[34]Perez A G, Benjamins V R. Overview of knowledge sharing and reuse components:Ontologies and problem-solving methods. Citeseer,1999
[35]宋炜,张铭.语义网简明教程高等教育出版社,2004.
[36]Bateman J A, Henschel R, Rinaldi F. The generalized upper model 2.0. On-line publication at I, 2005,1
[37]Bateman J A, Magnini B, Fabris G. The generalized upper model knowledge base:Organization and use. Towards very large knowledge bases:knowledge building and knowledge sharing,1995,: 60～72
[38]Miller G A. WordNet:a lexical database for English. Communications of the ACM,1995,38(11): 41
[39]Bylander T, Chandrasekaran B. Generic tasks for knowledge-based reasoning:The" right" level of abstraction for knowledge acquisition. INT. J. MAN MACH. STUD.,1987,26(2):231～243
[40]Chandrasekaran B. Generic tasks as building blocks for knowledge-based systems:The diagnosis and routine design examples. The Knowledge Engineering Review,2009,3(03):183～210
[41]Uschold M, King M, Moralee S, et al. The enterprise ontology. The knowledge engineering review,1998,13(01):31～89
[42]Antoniou G, Van Harmelen F. Web ontology language:Owl. Handbook on ontologies,2004,2: 45-60
[43]Corcho O, Fernandez-Lopez M, G6mez-Perez A. Methodologies, tools and languages for building ontologies. Where is their meeting point?. Data & Knowledge Engineering,2003,46(1):41～ 64
[44]周竞涛,王明微XML+RDF-实现Web数据基于语义的描述[EB/OL].2003,
[45]Fensel D, Horrocks I, Van Harmelen F, et al. OIL:Ontology infrastructure to enable the Semantic Web. IEEE intelligent systems,2001,16(2):38～45
[46]Horrocks I. DAML+OIL:a description logic for the semantic web. Bulletin of the Technical Committee on,2002,51:4
[47]Horrocks I, Patel-Schneider P F, Van Harmelen F. Reviewing the design of DAML+OIL:An ontology language for the semantic web. Menlo Park, CA; Cambridge, MA; London; AAAI Press; MIT Press; 1999,2002
[48]Fensel D, Van Harmelen F, Horrocks I.'OIL and DAML OIL:Ontology Languages for the Semantic Web. Towards the Semantic Web:ontology-driven knowledge management,2003,
[49]Genesereth M R, Fikes R E. Knowledge interchange format, version 3.0 reference manual.1992,
[50]Heflin J, Hendler J, Luke S. SHOE:A knowledge representation language for internet applications.1999,
[51]Luke S, Spector L, Rager D, et al. Ontology-based Web agents. ACM,1997
[52]Heflin J, Heflin J, Hendler J, et al. Towards the semantic web:knowledge representation in a dynamic, distributed environment.2001,
[53]Karp P D, Chaudhri V K, Thomere J. XOL:An XML-based ontology exchange language. Version 0.3, July,1999,3
[54]Motta E. An overview of the OCML modelling language. Citeseer,1998
[55]Tanasescu V, Domingue J, Cabral L. Ocml ontologies to xml schema lowering. Citeseer,2004
[56]Farquhar A, Fikes R, Rice J. The ontolingua server:A tool for collaborative ontology construction. International Journal of Human-Computers Studies,1997,46(6):707～727
[57]Lenat D B, Guha R V. Building Large Knowledge Based Systems:Representation and Inference in the CYC Project.1990. Boston, Masschusetts:Addison-Wesley,1990,
[58]Elkan C, Greinery R. Building large knowledge-based systems:representation and inference in the Cyc project.
[59]Macgregor R, Bates R. The LOOM knowledge representation language.1987,
[60]Macgregor R M. Inside the LOOM description classifier. ACM Sigart Bulletin,1991,2(3):92
[61]Cardoso J. The semantic web vision:Where are we?. IEEE Intelligent Systems,2007,22(5): 84～88
[62]李景.本体理论在文献检索系统中的应用研究北京图书馆出版社,2005.
[63]黄伟.本体构建与语义集成研究.2005,
[64]韩婕,向阳.本体构建研究综述.计算机应用与软件,2007,24(009)：21～23
[65]陈禹六IDEF建模分析和设计方法清华大学出版社,1999.
[66]鲁强,王智广,陈明.一种履带式本体构建模型.计算机应用,2008,28(005)：1183～1185
[67]鲁强,陈明,王智广.基于履带式本体构建方法的多代理模型研究.计算机应用,2009,(002)
[68]Kumazawa T, Saito O, Kozaki K, et al. Toward knowledge structuring of sustainability science based on ontology engineering. Sustainability Science,2009,4(1):99～116
[69]Sunagawa E, Kozaki K, Kitamura Y, et al. Organizing role-concepts in ontology development environment:Hozo. Citeseer,2005
[70]Kozaki K, Sunagawa E, Kitamura Y, et al. Fundamental consideration of role concepts for ontology evaluation. Proc. of EON2006 Edinburgh, United Kingdom,2006,
[71]Sunagawa E, Kozaki K, Kitamura Y, et al. Role organization model in Hozo. Managing Knowledge in a World of Networks,2006,:67～81
[72]Mizoguchi R, Sunagawa E, Kozaki K, et al. The model of roles within an ontology development tool:Hozo. Applied Ontology,2007,2(2):159～179
[73]Hitzler P, Kr Tzsch M, Parsia B, et al. OWL 2 Web Ontology Language Primer. W3C Recommendation. http://www.w3.org/TR/owl2-primer,2009,
[74]Motik B, Grau B C, Horrocks I, et al. Owl 2 web ontology language:Profiles. W3C Working Draft,2008,
[75]w3c.http://semanticweb.org/wiki/OWL_2:
[76]高志强,潘越,马力.语义Web原理及应用：北京：机械工业出版社,2009
[77]朱江,季林,许长朋.基于新知识观的本体云影模型.计算机与数字工程,2009,37(010)：58～60
[78]李德毅,孟海军,史雪梅.隶属云和隶属云发生器.计算机研究与发展,1995,32(006)：15～20
[79]杜益鸟,李德毅.基于云的概念划分及其在关联采掘上的应用.软件学报,2001,12(002)：196-203
[80]Guarino N. Concepts, attributes and arbitrary relations::Some linguistic and ontological criteria for structuring knowledge bases. Data & Knowledge Engineering,1992,8(3):249～261
[81]Guarino N. Some ontological principles for designing upper level lexical resources. Citeseer, 1998
[82]Loebe F. Abstract vs. social roles-a refined top-level ontological analysis. Citeseer,2005
[83]Sowa J F. Top-level ontological categories. International Journal of Human-Computer Studies, 1995,43(5-6):669-685
[84]Sowa J F. Knowledge representation:logical, philosophical, and computational foundationsMIT Press,2000.
[85]Varzi A C. Parts, wholes, and part-whole relations:The prospects of mereotopology. Data & Knowledge Engineering,1996,20(3):259～286
[86]王晓东,张小红,王靖,等.基于“角色”和“关系”的时间Ontology构建.河南师范大学学报：自然科学版,2008,36(001)：29～31
[87]赵波,解敏,夏幼明.基于本体的子类/父类关系的分类体系形式化表示.云南师范大学学报：自然科学版,2008,28(003)：27～30
[88]刘康,黄奇.语义网中重量级本体的设计.图书情报工作,2006,50(006)：42～45
[89]Varzi A C. On the boundary between mereology and topology. H lder-Pichler-Tempsky,1994
[90]Simons P. Part/whole Ⅱ:Mereology since 1900. Handbook of Metaphysics and Ontology. Philosophia, Munich,1991,:672～675
[91]Bremen G. Meronymic relationships:From classical mereology to complex part-whole relations. The Semantics of Relationships:An Interdisciplinary Perspective,2002,:35
[92]连莉,马军,雷景生,等Part-Whole关系的细分及性质分析.计算机工程,2006,32(017)：83～85
[93]Mizoguchi R, Ikeda M. Towards ontology engineering. JOURNAL-JAPANESE SOCIETY FOR ARTIFICIAL INTELLIGENCE,1998,13:9～10
[94]张素香,李瑞杰,罗锋,等.海洋生态动力学模型的研究进展.海洋湖沼通报,2006,4：121～127
[95]赵强.一个简单海洋生态系统动力学模型的伴随同化应用研究.2006,
[96]沈国英,施并章.海洋生态学科学出版社,2002.
[97]蔡福龙.海洋放射生态学：原子能出版社,1998
[98]Noy N F, Mcguinness D L. Ontology development 101:A guide to creating your first ontology. 2001,
[99]Mizoguchi R. Part 1:introduction to ontological engineering. New Generation Computing,2003, 21(4):365～384
[100]Miller G A, Beckwith R, Fellbaum C, et al. Introduction to wordnet:An on-line lexical database*. International Journal of lexicography,1990,3(4):235
[101]Rodriguez M A. Assessing semantic similarity among spatial entity classes.2000,
[102]Jianliang X, Xiaowei M. A Web-Based Ontology Evaluation System.2008
[103]王文海.胶州湾自然环境概述[J].海岸工程,1986,5(3)：18～24
[104]赵淑江.胶州湾生态系统主要生态因子的长期变化.2002,
[105]Dongfang Y, Zhenhui G, Peiyan S, et al. Mechanism of Nutrient Silicon and Water Temperature Influences on Phytoplankton Growth海洋通报：英文版,2006,8(002)：49～59
[106]Yang D F, Zhang J, Lu J B, et al. Examination of silicate limitation of primary production in Jiaozhou Bay, China. I. Silicate being a limiting factor of phytoplankton primary production. Chinese Journal of Oceanology and Limnology,2002,20:208～225
[107]杨东方,王凡,高振会,等.胶州湾浮游藻类生态现象Marine Sciences,2004,28(6):71
[108]Fikes R, Hayes P, Horrocks I. OWL-QL--a language for deductive query answering on the Semantic Web. Web Semantics:Science, Services and Agents on the World Wide Web,2004,2(1): 19～29
[109]Yang D, Gao Z, Chen Y, et al. Examination of silicate limitation of primary production in Jiaozhou Bay, North China. III. Judgment method, rules and uniqueness of nutrient limitation among N, P, and Si. Chinese Journal of Oceanology and Limnology,2003,21:114～133
[110]张均顺,沈志良.胶州湾营养盐结构变化的研究.海洋与湖沼,1997,28(005)：529～535
[111]Jian-Liang X U, Jing X, Yong L. A Query Optimization Strategy Based on Domain Ontology. 2009,:510～513
[112]李广建,黄昆.元搜索引擎及其主要技术.情报科学,2002,20(002)：175～179
[113]朱茂盛,王斌,程学旗.元搜索引擎及其实现.计算机工程,2002,28(011)：11～12
[114]王芳,张晓林.元搜索引擎：原理与利用.现代图书情报技术,1998,6：18～21
[115]Beigi M, Benitez A B, Chang S F. Metaseek:A content-based meta-search engine for images. Citeseer,1997
[116]Lawrence S, Giles L. Inquirus, the NECI meta search engine. Computer networks and ISDN systems,1998,30(1-7):95～105
[117]Joachims T. Optimizing search engines using clickthrough data. ACM,2002
[118]Aslam J A, Montague M. Bayes optimal metasearch:a probabilistic model for combining the results of multiple retrieval systems (poster session). ACM,2000
[119]Si L, Callan J. A semisupervised learning method to merge search engine results. ACM Transactions on Information Systems (TOIS),2003,21(4):491
[120]Vogt C C, Cottrell G W. Fusion via a linear combination of scores. Information Retrieval,1999, 1(3):151～173
[121]Lewis D, Hill M. The TREC-4 filtering track.1996
[122]崔舒寸,冯博琴.融合搜索引擎结果集的模糊积分算法.2006,
[123]Osiriski S, Weiss D. Carrot2:Design of a flexible and efficient web information retrieval framework. Springer-Verlag New York Inc,2005
[124]Weiss D. Carrot2 Developers Guide.
[125]Stefanowski J, Weiss D. Carrot 2 and language properties in web search results clustering. Advances in Web Intelligence,2003,:955
[126]许力,牟晓光,马云存.JPPF在构建高性能ETL过程中的应用.计算机应用,2008,28(z2)
[127]洪迎春.基于本体的分布式并行检索技术研究.2009,
