We describe improved mechanisms to accurately classify days when news for topics receive unexpectedly high amount of coverage. We further investigate the factors which influence this classification using ‘Presidential Elections’ as the topic of interest. This helps in bringing out useful trends and relations between days with hot topics by varying variables like history window size,van-ratio etc. We also propose a statistical scheme to approximate major events related to the topic. We then try to approximate the chain of events related to the major events. This can support a news alert service and also serve the purpose of automatically tracking news which follow up major events.
[1]
Jing Tao,et al.
Approach to extracting hot topics based on network traffic content
,
2009
.
[2]
Philip S. Yu,et al.
Parameter Free Bursty Events Detection in Text Streams
,
2005,
VLDB.
[3]
Ido Dagan,et al.
Keyword-Based Browsing and Analysis of Large Document Sets
,
1996
.
[4]
Haym Hirsh,et al.
Mining Associations in Text in the Presence of Background Knowledge
,
1996,
KDD.
[5]
Mark Wasson,et al.
Monitoring a newsfeed for hot topics
,
1999,
KDD '99.
[6]
Ido Dagan,et al.
Knowledge Discovery in Textual Databases (KDT)
,
1995,
KDD.