期刊导航

论文摘要

异构Web对象共指冲突识别与消解研究

Research on heterogeneous Web object coreference recognition and resolution algorithm

作者:王 珉(中国传媒大学计算机学院);王永滨(中国传媒大学计算机学院);李 樱(中国传媒大学计算机学院)

Author:WANG Min(School of Computer, Communication University of China);WANG Yong Bin(School of Computer, Communication University of China);LI Ying(School of Computer, Communication University of China)

收稿日期:2015-04-13          年卷(期)页码:2015,52(5):993-1000

期刊名称:四川大学学报: 自然科学版

Journal Name:Journal of Sichuan University (Natural Science Edition)

关键字:对象共指;冲突识别;冲突消解;规则推理;模式匹配

Key words:Object coreference recognition; Conflict; Conflict resolution; Rule based reasoning; Pattern matching

基金项目:国家科技支撑项目(2013BAH66F02)

中文摘要

针对现有的语义Web 中对象共指的消解工作研究不足,提出了一种异构Web数据流对象共指消解的技术.首先,定义了Web数据流对象的对象共指冲突识别与消解框架;然后,提出了一种基于描述逻辑的对象共指识别技术,归纳给出了描述逻辑的层次依赖关系并提出了基于依赖图的冲突遍历识别算法;并且,在冲突消解方面,定义了冲突的三种类型、冲突消解匹配模式与仲裁消解算法;最后,分别用测试集检验了本文提出的异构Web数据流冲突识别与消解技术的应用效果.

英文摘要

To resolve the coreference problem of the semantic Web data stream, this paper proposed a heterogeneous Web data stream object coreference resolution. First, the paper defined a framework to recognize and resolve Web data stream object coreference. Secondly, the paper described a conflict recognition system structure based on rule inference. The conflict recognition algorithm makes use of a fast rule matching, which improves the conflict recognition process efficiency and reduces the total refers to objects in space, time conflict and complexity of the digestion. In third, the paper defines three types of conflict and arbitration resolution algorithms for conflict resolution. Lastly, test sets are used in this paper to examine the application of heterogeneous Web data stream conflict recognition and coreference technology.

关闭

Copyright © 2020四川大学期刊社 版权所有.

地址:成都市一环路南一段24号

邮编:610065