Sketching Multidimensional Time Series for Fast Discord Mining
Not on the map. Why?
Abstract (arXiv)
Time series discords are a useful primitive for time series anomaly detection, and the matrix profile is capable of capturing discord effectively. There exist many research efforts to improve the scalability of discord discovery with respect to the length of time series. However, there is surprisingly little work focused on reducing the time complexity of matrix profile computation associated with dimensionality of a multidimensional time series. In this work, we propose a sketch for discord mining among multi-dimensional time series. After an initial pre-processing of the sketch as fast as reading the data, the discord mining has runtime independent of the dimensionality of the original data. On several real world examples from water treatment and transportation, the proposed algorithm improves the throughput by at least an order of magnitude (50X) and only has minimal impact on the quality of the approximated solution. Additionally, the proposed method can handle the dynamic addition or deletion of dimensions inconsequential overhead. This allows a data analyst to consider "what-if" scenarios in real time while exploring the data.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 13 of its 58 references that are in the map. First posted on arXiv 5 Nov 2023.
Builds on
13 of 58 references.
References · 13
- 2014Sketching as a Tool for Numerical Linear Algebracited by 285
- 2019MAD-GAN: Multivariate Anomaly Detection for Time Series Data with Generative Adversarial Networkscited by 176
- 2020Current Time Series Anomaly Detection Benchmarks are Flawed and are Creating the Illusion of Progresscited by 148
- 2018Temporal Pattern Attention for Multivariate Time Series Forecastingcited by 125
- 2021Towards a Rigorous Evaluation of Time-series Anomaly Detectioncited by 95
- 2018On Coresets for Logistic Regressioncited by 65
- 2015Frequent Directions : Simple and Deterministic Matrix Sketchingcited by 39
- 2019Discrepancy, Coresets, and Sketches in Machine Learningcited by 30
- 2022Embedding Compression with Hashing for Efficient Representation Learning in Large-Scale Graphcited by 10
- 2021Error-bounded Approximate Time Series Joins Using Compact Dictionary Representations of Time Seriescited by 9
- 2018Time is of the Essence: Machine Learning-based Intrusion Detection in Industrial Time Series Datacited by 6
- 2019Discovering Subdimensional Motifs of Different Lengths in Large-Scale Multivariate Time Seriescited by 3
Built on by
Citing papers per year: 2024: 2; 2025: 1.
3 papers.