Interactive time series exploration powered by the marriage of similarity distances

Abstract

Finding similar trends among time series data is critical for applications ranging from financial planning to policy making. The detection of these multifaceted relationships, especially time warped matching of time series of different lengths and alignments is prohibitively expensive to compute. To achieve real time responsiveness on large time series datasets, we propose a novel paradigm called Online Exploration of Time Series (ONEX) employing a powerful one-time preprocessing step that encodes critical similarity relationships to support subsequent rapid data exploration. Since the encoding of a huge number of pairwise similarity relationships for all variable lengths time series segments is not feasible, our work rests on the important insight that clustering with inexpensive point-to-point distances such as the Euclidean Distance can support subsequent time warped matching. Our ONEX framework overcomes the prohibitive computational costs associated with a more robust elastic distance namely the DTW by applying it over the surprisingly compact knowledge base instead of the raw data. Our comparative study reveals that ONEX is up to 19% more accurate and several times faster than the state-of-the-art. Beyond being a highly accurate and fast domain independent solution, ONEX offers a truly interactive exploration experience supporting novel time series operations.

Exploring change

“…Time series exploration. Exploration of time series data has been looked at by countless researchers [26], for instance to find similar time series [41,44]. These time series analysis techniques provide solutions for important parts of change exploration, such as obtaining stochastic models or identifying temporal update patterns.…”

Efficient Time Series Clustering by Minimizing Dynamic Time Warping Utilization

“…It can discover the underlying structure of the chaotic/raw datasets without the ground truth labels. This makes it particularly useful for analyzing many unlabeled real-world datasets, such as common pattern discovery [5], information retrieval [6] and outlier detection [7].…”

“…To significantly reduce the DTW utilization ratio for the acceleration, we only apply the complex DTW calculation on a summarized time series dataset (rather than the original dataset). This is inspired by the work [6] that achieves interactive time series retrieval by querying a summarized VOLUME 4, 2016 database, rather than the original large dataset. Specifically, we find L1-norm distance is effective to summarize the time series dataset based on three observations.…”

Interactive Time Series Analytics Powered by ONEX

“…We counter the prohibitively expensive use of DTW by employing a "marriage" of two distances: we use the computationally inexpensive Euclidean Distance to construct compact similarity groups for specific lengths, and then we support the exploration of these groups using the robust time-warping method DTW [2]. This unique combination yields very accurate results at much reduced response time rates, as DTW is successfully applied over the compact ONEX base instead of the raw data [5]. Our theoretical foundation rests on the proof of a triangle inequality between ED and DTW that builds a conceptual bridge between the offline construction of the ONEX base and its online exploration.…”

“…Our theoretical foundation rests on the proof of a triangle inequality between ED and DTW that builds a conceptual bridge between the offline construction of the ONEX base and its online exploration. Offering a novel answer to the trade-off dilemma between the use of complex time warped distances and timely responsiveness, ONEX has been shown to be several times faster than the fastest known method [6], while still delivering up to 19% more accurate results [5]. Our web-based visual analytics interface enable analysts to explore and directly interact with time series data sets in an intuitive manner using rich classes of operations.…”