PROJECT TITLE:

Efficient Motif Discovery for Large-scale Time Series in Healthcare - 2015

ABSTRACT:

Analyzing time series data will reveal the temporal behavior of the underlying mechanism manufacturing the info. Time series motifs, which are similar subsequences or frequently occurring patterns, have vital meanings for researchers especially in medical domain. With the fast growth of time series information, ancient ways for motif discovery are inefficient and not applicable to giant-scale data. This work proposes an economical Motif Discovery method for Large-scale time series (MDLats). By computing standard motifs, MDLats eliminates a majority of redundant computation in the connected arts and reuses existing information to the utmost. All the motif types and subsequences are generated for subsequent analysis and classification. Our system is implemented on a Hadoop platform and deployed during a hospital for clinical electrocardiography classification. The experiments on real-world healthcare information show that MDLats outperform the state-of-the-art methods even in massive time series.