首页|Events Detection and Temporal Analysis in Social Media

Events Detection and Temporal Analysis in Social Media

扫码查看
原文链接
  • NETL
  • Springer Nature
In the past few years， event detection has drawn a lot of attention。 We proposed an efficient method to detect event in this paper。 An event is defined as a set of descriptive， collocated keywords in this paper。 Intuitively， documents that describe the same event will contain similar sets of keywords。 Individual events will form clusters in the graph of keywords for a document collection。 We built a network of keywords based on their co-occurrence in documents。 We proposed an efficient method which create a keywords weight directed graph named KeyGraph and use community detection method to discover events。 Clump of keywords describing an event can be used to analyse the trend of the event。 The accuracy of detecting events is over eighty percents with our method。

Event detectionKeyGraphCo-occurrenceTemporal analysis

Yawei Jia、Jing Xu、Zhonghu Xu、Kai Xing

展开 >

University of Science and Technology of China, No. 443, Huangshan Road, Shushan District, Hefei 230027, Anhui, China

International conference on computer processing of oriental languages;CCF conference on natural language processing and Chinese computing

Kunming(CN)

Natural language understanding and intelligent applications

401-412

2016

10.1007/978-3-319-50496-4_33