Metric and algorithm for similarity between two temporal event sequences calculation
DOI:
https://doi.org/10.20535/SRIT.2308-8893.2017.3.12Keywords:
time stamped quasi periodic events, temporal event sequences similarity metric, computing distance between two real arrays of different length, accuracy, recall, precision for temporal event sequences, algorithm for distance of two event flows estimationAbstract
In several data analysis applications on temporal events flows, the problem of measuring "similarity" of these sequences arises. There are many different definitions of event sequences but in this paper under the term "sequence of events" the ordered array of the event occurrence times will be understood. In this paper the metric and procedure for similarity calculation between two ordered event sequences is presented. The procedure as the output returns measure of two event flows similarity and set of corresponding indices pairs which represent the mapping of the events between the input sequences.References
Pollock Gary. Holistic trajectories: a study of combined employment, housing and family careers by using multiple-sequence analysis. Journal of the Royal Statistical Society, Series A (Statistics in Society), 170(1): P. 167–183, 2007.
Mannila H., Moen P. Similarity between event types in sequences, Proc. First Intl. Conf. on Data Warehousing and Knowledge Discovery (DaWaK’99), Florence, Italy, 1999, P. 271–280.
Moen P., Attribute, Event Sequence, and Event Type Similarity Notions for Data Mining, Ph.D. thesis, Department of Computer Science, University of Helsinki, Finland, 2000.