An efficient pattern mining approach for event detection in multivariate temporal data. Issue 1 (January 2016)

• Record Type: Journal Article
• Authors: Batal, Iyad; Cooper, Gregory; Fradkin, Dmitriy; Harrison, James; Moerchen, Fabian; Hauskrecht, Milos
• Is Part Of: Knowledge and information systems. Volume 46:Issue 1(2016:Jan.)
• Page Start: 115
• Page End: 150
• Publication Date: 2016-01
• Subjects: Temporal data mining -- Electronic health records -- Temporal abstractions -- Time-interval patterns -- Recent temporal patterns -- Event detection; Expert systems (Computer science) -- Periodicals; Information storage and retrieval systems -- Periodicals; 006.33
• DOI: 10.1007/s10115-015-0819-6
• Languages: English
• ISSNs: 0219-1377

Abstract

Abstract This work proposes a pattern mining approach to learn event detection models from complex multivariate temporal data, such as electronic health records. We present recent temporal pattern mining, a novel approach for efficiently finding predictive patterns for event detection problems. This approach first converts the time series data into time-interval sequences of temporal abstractions. It then constructs more complex time-interval patterns backward in time using temporal operators. We also present the minimal predictive recent temporal patterns framework for selecting a small set of predictive and non-spurious patterns. We apply our methods for predicting adverse medical events in real-world clinical data. The results demonstrate the benefits of our methods in learning accurate event detection models, which is a key step for developing intelligent patient monitoring and decision support systems.