Finding unusual medical time-series subsequences: Algorithms and applications

E Keogh, J Lin, AW Fu… - IEEE Transactions on …, 2006 - ieeexplore.ieee.org
E Keogh, J Lin, AW Fu, H Van Herle
IEEE Transactions on Information Technology in Biomedicine, 2006ieeexplore.ieee.org
In this work, we introduce the new problem of finding time series discords. Time series
discords are subsequences of longer time series that are maximally different to all the rest of
the time series subsequences. They thus capture the sense of the most unusual
subsequence within a time series. While discords have many uses for data mining, they are
particularly attractive as anomaly detectors because they only require one intuitive
parameter (the length of the subsequence), unlike most anomaly detection algorithms that …
In this work, we introduce the new problem of finding time series discords. Time series discords are subsequences of longer time series that are maximally different to all the rest of the time series subsequences. They thus capture the sense of the most unusual subsequence within a time series. While discords have many uses for data mining, they are particularly attractive as anomaly detectors because they only require one intuitive parameter (the length of the subsequence), unlike most anomaly detection algorithms that typically require many parameters. While the brute force algorithm to discover time series discords is quadratic in the length of the time series, we show a simple algorithm that is three to four orders of magnitude faster than brute force, while guaranteed to produce identical results. We evaluate our work with a comprehensive set of experiments on electrocardiograms and other medical datasets
ieeexplore.ieee.org