语义蚁群算法的不确定知识本体推理研究
|
摘要
语义网最初被认为是现在万维网的一个延伸，它可以帮助计算机理解网络中的信息，从而可以支持更加复杂的发现、数据整合、导航以及任务的自动分配。语义网背后最主要的思想就是给网页上添加可以使机器理解的含义，通过使用本体技术为网络资源的共享术语定义精确的含义，利用知识表示技术来进行网络资源的自动推理，并且使用协作代理技术来处理网络上的信息。不确定描述逻辑知识库扩展了传统的描述逻辑知识库，特别是引入了关于概念和角色的不确定知识，同时也引入了关于概念和角色个体的不确定知识。前者是关于随机选到的概念和角色实例的不确定知识，后者是关于离散的概念和角色实例的不确定知识。实际应用迫切需要处理本体中的不确定知识，特别是在医药、生物、天文学等领域，不确定描述逻辑知识可以比较好的运用在这些领域。
在不确定描述逻辑关键性问题中，有一大类问题是关于求解公理的概率上限和下限，其实质是一个求极值问题，考虑到今后发展的趋势，这种极值问题很有可能会随着不确定描述逻辑的深入研究，发展成为非线性多维度极值问题。因而，本文在蚁群算法的基础上，基于不确定描述逻辑领域的特征，提出了一种基于语义信息的蚁群算法，进而将其应用于求解上述的极值问题，并讨论其关键性参数的取值，最后，通过一个医学本体案例，验证了本文提出的基于语义信息的蚁群算法的有效性。
Semantic network is considered as extension of World Wide Web. It can help thecomputer understand the network information, which can support much more complex dataintegration, navigation, and so. The main ideas behind the Semantic Web are to add meaning,which can make machines understand, into the web page. Semantic network containsuncertainty knowledge which is distributed in different nodes of network. This problem canbe solved by the uncertain knowledge of data integration technology in order to depict thetrue extent. This uncertain knowledge of data integration can produce uncertain knowledge ofthe facts.
The uncertainty description logic knowledge base extends the traditional descriptionlogic knowledge base, especially with the introduction of uncertain knowledge aboutconcepts and roles, also the uncertainty knowledge about concepts and roles of individuals.The former is randomly selected, while the latter is discrete. The practical applications areurgent to deal with uncertain knowledge in the Ontology, especially in areas such as medicine,biology, astronomy and so on.
As to traditional description logic reasoning, there are classic determine ontologyreasoning logics, such as SHIF (D) and SHOIN (D). They are corresponding to the networkontology language OWL Lite and OWL DL, and also have some successful applications.However, little actual projects are done for the uncertainty logic description reasoningalthough this kind of logic reasoning has the uncertain knowledge ontology reasoning logic.
One of the uncertain logic description key issues is the method to calculate the upper andlower limits of the axiom of probability. In this paper, a improved ant colony algorithm basedon semantic ontology is proposed to meet this need. Finally, by a medical ontology case, thisalgorithm’s validity are verified.
在不确定描述逻辑关键性问题中，有一大类问题是关于求解公理的概率上限和下限，其实质是一个求极值问题，考虑到今后发展的趋势，这种极值问题很有可能会随着不确定描述逻辑的深入研究，发展成为非线性多维度极值问题。因而，本文在蚁群算法的基础上，基于不确定描述逻辑领域的特征，提出了一种基于语义信息的蚁群算法，进而将其应用于求解上述的极值问题，并讨论其关键性参数的取值，最后，通过一个医学本体案例，验证了本文提出的基于语义信息的蚁群算法的有效性。
Semantic network is considered as extension of World Wide Web. It can help thecomputer understand the network information, which can support much more complex dataintegration, navigation, and so. The main ideas behind the Semantic Web are to add meaning,which can make machines understand, into the web page. Semantic network containsuncertainty knowledge which is distributed in different nodes of network. This problem canbe solved by the uncertain knowledge of data integration technology in order to depict thetrue extent. This uncertain knowledge of data integration can produce uncertain knowledge ofthe facts.
The uncertainty description logic knowledge base extends the traditional descriptionlogic knowledge base, especially with the introduction of uncertain knowledge aboutconcepts and roles, also the uncertainty knowledge about concepts and roles of individuals.The former is randomly selected, while the latter is discrete. The practical applications areurgent to deal with uncertain knowledge in the Ontology, especially in areas such as medicine,biology, astronomy and so on.
As to traditional description logic reasoning, there are classic determine ontologyreasoning logics, such as SHIF (D) and SHOIN (D). They are corresponding to the networkontology language OWL Lite and OWL DL, and also have some successful applications.However, little actual projects are done for the uncertainty logic description reasoningalthough this kind of logic reasoning has the uncertain knowledge ontology reasoning logic.
One of the uncertain logic description key issues is the method to calculate the upper andlower limits of the axiom of probability. In this paper, a improved ant colony algorithm basedon semantic ontology is proposed to meet this need. Finally, by a medical ontology case, thisalgorithm’s validity are verified.
引文
[1] Berners-Lee, T., J. Hendler, and O. Lassila, The Semantic Web-A new form of Webcontent that is meaningful to computers will unleash a revolution of new possibilities.Scientific American[J],2001.284(5):34.
[2] W3C. OWL web ontology language overview.2004. W3C Recommendation[EB/OL];Available from: http://www.w3.org/TR/2004/REC-owl-features-20040210/,2004.
[3] Horrocks, I., P.F. Patel-Schneider, and F. Van Harmelen, From SHIQ and RDF to OWL:The making of a Web Ontology Language[J]. Web Semantics,2003.1(1):7-26.
[4] Horrocks, I. and P.F. Patel-Schneider, Reducing OWL entailment to description logicsatisfiability[J]. Semantic Web-Iswc2003,2003.2870:17-29.
[5] F. Baader, D.C., D. McGuinness, D. Nardi, P.F. Patel-Schneider (Eds.), The DescriptionLogic Handbook: Theory, Implementation,and Applications[M].2003: CambridgeUniversity Press.
[6] Lukasiewicz, T., Fuzzy description logic programs under the answer set semantics for thesemantic web[C]. RuleML2006: Second International Conference on Rules and RuleMarkup Languages for the Semantic Web, Proceedings,2006:89-96.
[7] Lukasiewicz, T., Probabilistic description logic programs. International Journal ofApproximate Reasoning[J],2007.45(2):288-307.
[8] Straccia, U., Fuzzy description logic programs[C], in Proceedings IPMU-2006.2006:1818–1825.
[9] Costa, P.C.G. and K.B. Laskey, PR-OWL: A Framework for Probabilistic Ontologies[J].Formal Ontology in Information Systems,2006.150:237-249.
[10]Hung, E., Y. Deng, and V.S. Subrahmanian. TOSS: An extension of TAX with ontologiesand similarity queries[C]. in Proceedings of the ACM SIGMOD International Conferenceon Management of Data, SIGMOD2004, Jun13-182004.2004. Paris, France:Association for Computing Machinery.
[11] Udrea, O., et al., Probabilistic ontologies and relational databases[C]. On the Move toMeaningful Internet Systems2005: Coopis, Doa, and Odbase, Pt1, Proceedings,2005.3760:1-17.
[12]Mitra, P., N.F. Noy, and A.R. Jaiswal, OMEN: A probabilistic ontology mapping tool[C].Semantic Web-Iswc2005, Proceedings,2005.3729:537-547.
[13]Pan, R., et al., A Bayesian network approach to ontology mapping[J]. Semantic Web-Iswc2005, Proceedings,2005.3729:563-577.
[14]Halevy, A., M. Franklin, and D. Maier. Principles of dataspace systems[C]. in25th ACMSIGMOD-SIGACT-SIGART Symposium on Principles of Database Systems, PODS2006, June26,2006-June28,2006.2006. Chicago, IL, United states: Association forComputing Machinery.
[15]van Keulen, M., A. de Keijzer, and W. Alink, A probabilistic XML approach to dataintegration[C]. Icde2005:21st International Conference on Data Engineering,Proceedings,2005:459-470.
[16]Richardson, M. and P. Domingos, Markov logic networks[J]. Machine Learning,2006.62(1-2):107-136.
[17]I. Horrocks, U.S., S. Tobies,. Practical reasoning for expressive description logics[C]. inProceedings LPAR-1999.1999. LNCS: Springer.
[18]Horrocks, I. and U. Sattler, A Tableaux Decision Procedure for SHOTQ[C].19thInternational Joint Conference on Artificial Intelligence (Ijcai-05),2005:448-453.
[19]Lukasiewicz, T., Probabilistic deduction with conditional constraints over basic events[J].Journal of Artificial Intelligence Research,1999.10:199-241.
[20]Lukasiewicz, T., Probabilistic default reasoning with conditional constraints[J]. Annals ofMathematics and Artificial Intelligence,2002.34(1-3):35-88.
[21]Lukasiewicz, T., Probabilistic description logic programs under inheritance withoverriding for the Semantic Web[J]. International Journal of Approximate Reasoning,2008.49(1):18-34.
[22]Hailperin, T., Boole's Logic and Probability, second ed[M].1986: North-Holland.
[23]Hackwood, S. and G. Beni. Self-organization of sensors for swarm intelligence[C]. inProceedings1992IEEE International Conference on Robotics and Automation, May12,1992-May14,1992.1992. Nice, Fr: Publ by IEEE.
[24]Bonabeau.E, D.M., Theraulaz.G, Swarm intelligence from natural to artificial systems[M].1999: Oxford University Press
[25]张铃，程军盛,松散的脑袋一群体智能的数学模型.模式识别与人工智能,2003(01):1-5.
[26]葛芬,吴.,基于多种技术的Word设计文档自动生成平台[J].电子科技大学学报,2007.36(2):263-266.
[27]艾伟，孙四明，张峰,基于本体的Web文本挖掘与信息检索[J].计算机工程,2010.36(22):75-80.
[28]Arpirez, J.C., et al. WebODE: A scalable workbench for ontological engineering[J]. inProceedings of the First International Conference on Knowledge Capture, October21,2001-October23,2001.2001. Victoria, BC, Canada: Association for ComputingMachinery.
[29]高尚，杨静宇,群智能算法及其应用[M].2006:水利水电出版社,1-10.
[30]纪震,周.,廖惠连,吴青华,智能单粒子优化算法.计算机学报[J],2010.33(3):556-561.
[31]Chu, K.K., C.I. Lee, and R.S. Tsai, Ontology technology to assist learners' navigation inthe concept map learning system[J]. Expert Systems with Applications,2011.38(9):11293-11299.
[32]Zhao, C., et al. An ontology-based semantic search model study[J]. in20103rdInternational Symposium on Knowledge Acquisition and Modeling, KAM2010, October20,2010-October21,2010.2010. Wuhan, China: IEEE Computer Society.
[33]黄翰,郝.,吴春国,秦勇,蚁群算法的收敛速度分析[J].计算机学报,2007.30(8):1344-1353.
[34]张明明;秦永彬,基于前序关系的非确定型有穷自动机极小化算法[J].山东大学学报(理学版),2010.07:34-38.
[35]美国国家医学图书馆. ULMs Program [EB/OL], http://www.nlm.nih.gov/research/umls/.
[36]Thomas Lukasiewicz, Expressive probabilistic description logics[J]. ArtifcialIntelligence,2008.172:852–883.
[2] W3C. OWL web ontology language overview.2004. W3C Recommendation[EB/OL];Available from: http://www.w3.org/TR/2004/REC-owl-features-20040210/,2004.
[3] Horrocks, I., P.F. Patel-Schneider, and F. Van Harmelen, From SHIQ and RDF to OWL:The making of a Web Ontology Language[J]. Web Semantics,2003.1(1):7-26.
[4] Horrocks, I. and P.F. Patel-Schneider, Reducing OWL entailment to description logicsatisfiability[J]. Semantic Web-Iswc2003,2003.2870:17-29.
[5] F. Baader, D.C., D. McGuinness, D. Nardi, P.F. Patel-Schneider (Eds.), The DescriptionLogic Handbook: Theory, Implementation,and Applications[M].2003: CambridgeUniversity Press.
[6] Lukasiewicz, T., Fuzzy description logic programs under the answer set semantics for thesemantic web[C]. RuleML2006: Second International Conference on Rules and RuleMarkup Languages for the Semantic Web, Proceedings,2006:89-96.
[7] Lukasiewicz, T., Probabilistic description logic programs. International Journal ofApproximate Reasoning[J],2007.45(2):288-307.
[8] Straccia, U., Fuzzy description logic programs[C], in Proceedings IPMU-2006.2006:1818–1825.
[9] Costa, P.C.G. and K.B. Laskey, PR-OWL: A Framework for Probabilistic Ontologies[J].Formal Ontology in Information Systems,2006.150:237-249.
[10]Hung, E., Y. Deng, and V.S. Subrahmanian. TOSS: An extension of TAX with ontologiesand similarity queries[C]. in Proceedings of the ACM SIGMOD International Conferenceon Management of Data, SIGMOD2004, Jun13-182004.2004. Paris, France:Association for Computing Machinery.
[11] Udrea, O., et al., Probabilistic ontologies and relational databases[C]. On the Move toMeaningful Internet Systems2005: Coopis, Doa, and Odbase, Pt1, Proceedings,2005.3760:1-17.
[12]Mitra, P., N.F. Noy, and A.R. Jaiswal, OMEN: A probabilistic ontology mapping tool[C].Semantic Web-Iswc2005, Proceedings,2005.3729:537-547.
[13]Pan, R., et al., A Bayesian network approach to ontology mapping[J]. Semantic Web-Iswc2005, Proceedings,2005.3729:563-577.
[14]Halevy, A., M. Franklin, and D. Maier. Principles of dataspace systems[C]. in25th ACMSIGMOD-SIGACT-SIGART Symposium on Principles of Database Systems, PODS2006, June26,2006-June28,2006.2006. Chicago, IL, United states: Association forComputing Machinery.
[15]van Keulen, M., A. de Keijzer, and W. Alink, A probabilistic XML approach to dataintegration[C]. Icde2005:21st International Conference on Data Engineering,Proceedings,2005:459-470.
[16]Richardson, M. and P. Domingos, Markov logic networks[J]. Machine Learning,2006.62(1-2):107-136.
[17]I. Horrocks, U.S., S. Tobies,. Practical reasoning for expressive description logics[C]. inProceedings LPAR-1999.1999. LNCS: Springer.
[18]Horrocks, I. and U. Sattler, A Tableaux Decision Procedure for SHOTQ[C].19thInternational Joint Conference on Artificial Intelligence (Ijcai-05),2005:448-453.
[19]Lukasiewicz, T., Probabilistic deduction with conditional constraints over basic events[J].Journal of Artificial Intelligence Research,1999.10:199-241.
[20]Lukasiewicz, T., Probabilistic default reasoning with conditional constraints[J]. Annals ofMathematics and Artificial Intelligence,2002.34(1-3):35-88.
[21]Lukasiewicz, T., Probabilistic description logic programs under inheritance withoverriding for the Semantic Web[J]. International Journal of Approximate Reasoning,2008.49(1):18-34.
[22]Hailperin, T., Boole's Logic and Probability, second ed[M].1986: North-Holland.
[23]Hackwood, S. and G. Beni. Self-organization of sensors for swarm intelligence[C]. inProceedings1992IEEE International Conference on Robotics and Automation, May12,1992-May14,1992.1992. Nice, Fr: Publ by IEEE.
[24]Bonabeau.E, D.M., Theraulaz.G, Swarm intelligence from natural to artificial systems[M].1999: Oxford University Press
[25]张铃，程军盛,松散的脑袋一群体智能的数学模型.模式识别与人工智能,2003(01):1-5.
[26]葛芬,吴.,基于多种技术的Word设计文档自动生成平台[J].电子科技大学学报,2007.36(2):263-266.
[27]艾伟，孙四明，张峰,基于本体的Web文本挖掘与信息检索[J].计算机工程,2010.36(22):75-80.
[28]Arpirez, J.C., et al. WebODE: A scalable workbench for ontological engineering[J]. inProceedings of the First International Conference on Knowledge Capture, October21,2001-October23,2001.2001. Victoria, BC, Canada: Association for ComputingMachinery.
[29]高尚，杨静宇,群智能算法及其应用[M].2006:水利水电出版社,1-10.
[30]纪震,周.,廖惠连,吴青华,智能单粒子优化算法.计算机学报[J],2010.33(3):556-561.
[31]Chu, K.K., C.I. Lee, and R.S. Tsai, Ontology technology to assist learners' navigation inthe concept map learning system[J]. Expert Systems with Applications,2011.38(9):11293-11299.
[32]Zhao, C., et al. An ontology-based semantic search model study[J]. in20103rdInternational Symposium on Knowledge Acquisition and Modeling, KAM2010, October20,2010-October21,2010.2010. Wuhan, China: IEEE Computer Society.
[33]黄翰,郝.,吴春国,秦勇,蚁群算法的收敛速度分析[J].计算机学报,2007.30(8):1344-1353.
[34]张明明;秦永彬,基于前序关系的非确定型有穷自动机极小化算法[J].山东大学学报(理学版),2010.07:34-38.
[35]美国国家医学图书馆. ULMs Program [EB/OL], http://www.nlm.nih.gov/research/umls/.
[36]Thomas Lukasiewicz, Expressive probabilistic description logics[J]. ArtifcialIntelligence,2008.172:852–883.