Paper Info
Title
Discovering event episodes from sequences of online news articles: A time-adjoining frequent itemset-based clustering method
Abstract
•We develop a novel time-adjoining frequent itemset-based method to discover event episodes from sequences of online news articles.•The proposed method incorporates temporal characteristics of news articles that reveal distinct episodes of an event.•We use 1,468 online news articles that pertain to 248 episodes and 53 events to evaluate the proposed method and several benchmarks.•The results show that our method outperforms the benchmarks consistently and significantly as measured by cluster recall and cluster precision.•We further analyze the effect of temporal proximity on our method’s performance for discovering event episodes from news articles.
Year
DOI
Venue
2020
10.1016/j.im.2020.103348
Information & Management
Keywords
DocType
Volume
Event episode discovery,Retrospective event detection,Event evolution,Temporal frequent itemset-based clustering
Journal
57
Issue
ISSN
Citations 
7
0378-7206
0
PageRank 
References 
Authors
0.34
0
4
Name
Order
Citations
PageRank
Yen-hsien Lee111816.64
Paul Jen-hwa Hu22046112.56
Hongquan Zhu300.34
Hsin-Wei Chen400.34