Paper Info
Title
Topic dynamics: an alternative model of bursts in streams of topics
Abstract
For some time there has been increasing interest in the problem of monitoring the occurrence of topics in a stream of events, such as a stream of news articles. This has led to different models of bursts in these streams, i.e., periods of elevated occurrence of events. Today there are several burst definitions and detection algorithms, and their differences can produce very different results in topic streams. These definitions also share a fundamental problem: they define bursts in terms of an arrival rate. This approach is limiting; other stream dimensions can matter. We reconsider the idea of bursts from the standpoint of a simple kind of physics. Instead of focusing on arrival rates, we reconstruct bursts as a dynamic phenomenon, using kinetics concepts from physics -- mass and velocity -- and derive momentum, acceleration, and force from these. We refer to the result as topic dynamics, permitting a hierarchical, expressive model of bursts as intervals of increasing momentum. As a sample application, we present a topic dynamics model for the large PubMed/MEDLINE database of biomedical publications, using the MeSH (Medical Subject Heading) topic hierarchy. We show our model is able to detect bursts for MeSH terms accurately as well as efficiently.
Year
DOI
Venue
2010
10.1145/1835804.1835862
KDD
Keywords
Field
DocType
different model,topic dynamics model,mesh term,stream dimension,expressive model,alternative model,derive momentum,topic stream,topic dynamic,topic hierarchy,arrival rate,hierarchy,momentum,kinetics,subject headings
Data mining,Computer science,Acceleration,Momentum,Artificial intelligence,STREAMS,Hierarchy,Machine learning,Limiting
Conference
Citations 
PageRank 
References 
41
1.58
17
Authors
2
Name
Order
Citations
PageRank
Dan He113312.54
D. Stott Parker2687182.27