首页> 外文会议>International Conference on Knowledge Science, Engineering and Management >Micro-blog Hot Event Detection Based on Dynamic Event Model
【24h】

Micro-blog Hot Event Detection Based on Dynamic Event Model

机译:基于动态事件模型的微博热事件检测

获取原文

摘要

Micro-blog is a rapid developing platform, where users can share information about the event happened in the real life, and sometimes many hot events even originate in this platform. So it is very important to detect the hot event in the micro-blog texts. Micro-blog text is a kind of the short, fractional and grass-root text. Based on the analysis of the dynamic evolvement property of the event, we propose a micro-blog event detection method based on dynamic event model, where the event model is created based on the two centroids. This paper firstly analyzes the dynamic evolvement property, secondly introduces the creation and modification process of the two-centriods based event model, and finally gives the event detection method based on the proposed dynamic event model. Experimental results showed that our proposed model improves the performances of the micro-blog event detection greatly.
机译:Micro-Blog是一个快速发展的平台,用户可以分享有关现实生活中发生的事件的信息,有时许多热门事件甚至源于这个平台。因此,在微博文本中检测到热门事件非常重要。微博文本是一种短,分数和草根文本。基于对事件的动态演进性质的分析,我们提出了一种基于动态事件模型的微博事件检测方法,其中基于两个质心创建事件模型。本文首先分析了动态演进性,其次介绍了基于二厘芯的事件模型的创建和修改过程,最后给出了基于所提出的动态事件模型的事件检测方法。实验结果表明,我们的拟议模型大大提高了微博事件检测的性能。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号