面向网络大数据的知识融合方法综述
发布时间：2018-07-22 17:22
【摘要】：网络大数据是指"人、机、物"三元世界在网络空间中交互、融合所产生并在互联网上可获得的大数据.网络大数据中蕴含丰富的知识资源,包括描述特定事物的实体、刻画实体逻辑联系的关系、用于语义标注实体的分类等.知识自身呈现出异质性、多元性和碎片化等特点.如何在网络大数据环境下海量碎片化的数据中提取出能够用于解决问题的知识,并对知识进行有效的融合计算,将从网络大数据中获得的知识有效组织起来是知识库构建亟待解决的技术难点和当前研究的热点.该文从知识融合的定义出发,介绍近年来的可用于知识融合的技术和算法的最新进展,通过分类和总结现有技术,为进一步的研究工作提供可选方案.文中首先介绍了在知识融合中用于判断知识真伪的知识评估的若干研究和评估方法;然后基于知识评估的结果,从实体扩充、关系扩充和分类扩充3个方面详细总结了知识融合中各种可用的知识扩充方法和研究进展;探讨了应用于网络大数据的知识融合的总体框架;基于这些讨论,总结面向网络大数据的知识融合面临的主要挑战和可能解决方案,并展望了该技术未来的发展方向与前景.
[Abstract]:Network big data refers to the big data generated by the interaction and fusion of "people, machines, and objects" in cyberspace and available on the Internet. The network big data contains abundant knowledge resources, including describing the entities of specific things, characterizing the relations of entity logical relations, and classifying entities for semantic annotation. Knowledge itself presents the characteristics of heterogeneity, pluralism and fragmentation. How to extract the knowledge which can be used to solve the problem from the massive fragmentation data in the network big data environment, and carry on the effective fusion computation to the knowledge, Effectively organizing the knowledge gained from the network big data is the technical difficulty and the hot spot of the current research to construct the knowledge base. Starting from the definition of knowledge fusion, this paper introduces the latest progress of the technologies and algorithms available for knowledge fusion in recent years. By classifying and summarizing the existing technologies, it provides an alternative scheme for further research work. This paper first introduces some research and evaluation methods of knowledge assessment used in knowledge fusion to judge knowledge authenticity, and then extends from entity based on the results of knowledge assessment. In this paper, three aspects of relationship expansion and classification expansion are summarized in detail, and the available methods and research progress of knowledge expansion in knowledge fusion are summarized in detail, the general framework of knowledge fusion applied to network big data is discussed, and based on these discussions, This paper summarizes the main challenges and possible solutions of knowledge fusion for network big data, and looks forward to the future development direction and prospect of the technology.
【作者单位】： 中国科学院信息工程研究所;中国科学院计算技术研究所;
【基金】：国家科技支撑计划(2012BAH46B03) 核高基项目(2013ZX01039-002-001-001) 国家重点研发计划(2016YFB1000902) 国家自然科学基金(61602467,61303056,61402442,61402464,61572469,61572473,61502478) 北京市自然科学基金项目(4154086)资助~~
【分类号】：TP311.13
，
本文编号：2138197
[Abstract]:Network big data refers to the big data generated by the interaction and fusion of "people, machines, and objects" in cyberspace and available on the Internet. The network big data contains abundant knowledge resources, including describing the entities of specific things, characterizing the relations of entity logical relations, and classifying entities for semantic annotation. Knowledge itself presents the characteristics of heterogeneity, pluralism and fragmentation. How to extract the knowledge which can be used to solve the problem from the massive fragmentation data in the network big data environment, and carry on the effective fusion computation to the knowledge, Effectively organizing the knowledge gained from the network big data is the technical difficulty and the hot spot of the current research to construct the knowledge base. Starting from the definition of knowledge fusion, this paper introduces the latest progress of the technologies and algorithms available for knowledge fusion in recent years. By classifying and summarizing the existing technologies, it provides an alternative scheme for further research work. This paper first introduces some research and evaluation methods of knowledge assessment used in knowledge fusion to judge knowledge authenticity, and then extends from entity based on the results of knowledge assessment. In this paper, three aspects of relationship expansion and classification expansion are summarized in detail, and the available methods and research progress of knowledge expansion in knowledge fusion are summarized in detail, the general framework of knowledge fusion applied to network big data is discussed, and based on these discussions, This paper summarizes the main challenges and possible solutions of knowledge fusion for network big data, and looks forward to the future development direction and prospect of the technology.
【作者单位】： 中国科学院信息工程研究所;中国科学院计算技术研究所;
【基金】：国家科技支撑计划(2012BAH46B03) 核高基项目(2013ZX01039-002-001-001) 国家重点研发计划(2016YFB1000902) 国家自然科学基金(61602467,61303056,61402442,61402464,61572469,61572473,61502478) 北京市自然科学基金项目(4154086)资助~~
【分类号】：TP311.13
，
本文编号：2138197
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2138197.html
