Discovery of Periodic Patterns in Sequence Data: A Variance-based Approach
INFORMS Journal on Computing, 2011
In this paper, Assistant Professor Catherine Yang and co-authors Balaji Padmanabhan from the University of Southern Florida and Hongyan Liu and Xiaoyu Wang from Tsinghua University address the discovery of periodic patterns in sequence data. Building on prior work in this area, the authors present definitions and new methods for characterizing and identifying four types of periodic patterns.
A unifying concept across the different types of periodic patterns the authors consider is the use of statistical variance to define periodicity. This lends itself to efficient variance-reduction algorithms for identifying periodic patterns. The authors motivate and test this approach using both extensive simulated sequences and real sequence data from online clickstream data.