Zhang, Dongzhi, Lee, Kyungmi, and Lee, Ickjai (2018) Mining medical periodic patterns from spatio-temporal trajectories. In: Lecture Notes in Computer Science (11148) pp. 123-133. From: HIS 2018: 7th International Conference on Health Information Science, 5-7 October 2018, Cairns, QLD, Australia.

PDF (Published Version) - Published Version Restricted to Repository staff only

 

2

1

Abstract

A spatio-temporal trajectory captures the movement behaviors of an object, and reveals various periodic patterns for the object such as where and when the object regularly visits. Due to the recent advances in GPS-enabled data collection devices such as mobile phones, a large set of spatio-temporal trajectories has been collected and available for analysis. These spatio-temporal trajectories could be used to identify those people who periodically visit medical centres for treatments (patients), working (health professionals) or other purposes. Spatio-temporal periodic pattern mining is to find periodic patterns for a certain place at regular intervals from spatio-temporal trajectories. Past studies attempt to find periodic patterns in medical contexts through time-series datasets, but not from spatio-temporal trajectories. In this study, we introduce a medical periodic pattern mining framework that utilises spatio-temporal periodic pattern mining approaches to find medical periodic patterns. We test the feasibility and applicability of our framework through a real-world publicly available dataset. Experimental results reveal that our framework is able to identify those people who regularly visit medical centres from those not, and also find medical periodic patterns revealing interesting medical behaviors.

Actions (Repository Staff Only)

Item Control Page