Anomaly Detection Paradigm for Multivariate Time Series Data Mining for Healthcare
Source abstract
Abstract
Time series data are significant, and are derived from temporal data, which involve real numbers representing values collected regularly over time. Time series have a great impact on many types of data. However, time series have anomalies. We introduce an anomaly detection paradigm called novel matrix profile (NMP) to solve the all-pairs similarity search problem for time series data in the healthcare. The proposed paradigm inherits the features from two state-of-the-art algorithms: Scalable Time series Anytime Matrix Profile (STAMP) and Scalable Time-series Ordered-search Matrix Profile (STOMP). The proposed NMP caches the output in an easy-to-access fashion for single- and multidimensional data. The proposed NMP can be used on large multivariate data sets and generates approximate solutions of high quality in a reasonable time. It is implemented on a Python platform. To determine its effectiveness, it is compared with the state-of-the-art matrix profile algorithms, i.e., STAMP and STOMP. The results confirm that the proposed NMP provides higher accuracy than the compared algorithms.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Time Series Analysis and Forecasting · Anomaly Detection Techniques and Applications · Traditional Chinese Medicine Studies
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Cyber Security, International Information Technology University, Almaty 050000, Kazakhstan
Dahaa Research Group, Department of Computer Science, Shaqra University, Shaqra 11961, Saudi Arabia
Sensor Networks and Cellular Systems (SNCS) Research Center, University of Tabuk, Tabuk 47731, Saudi Arabia
Department of Information Technology, University of Tabuk, Tabuk 47731, Saudi Arabia
Computer and Information Science College, Jouf University, Sakakah 72388, Saudi Arabia
Department of Electical and Computer Engineering, University of Arizona, Tucson, AZ 85721, USA
Computers and Information Technology College, Taif University, Taif 21974, Saudi Arabia
Funding
Sensors Networks and Cellular Systems (SNCS) Research Center
References (34)
Show source-provided references
1. Time works well: Dynamic time warping based on time weighting for time series data mining10.1016/j.ins.2020.08.089 ↗
2. Sattari, M.T., Avram, A., Apaydin, H., and Matei, O. (2020). Soil Temperature Estimation with Meteorological Parameters by Using Tree-Based Hybrid Data Mining Models. Mathematics, 8.10.3390/math8091407 ↗
3. Zhang, S.Q., and Zhou, Z.H. (September, January 29). Harmonic recurrent process for time series forecasting. Proceedings of the European Conference on Artificial Intelligence, Santiago de Compostela, Spain.
4. DLCSS: A new similarity measure for time series data mining10.1016/j.engappai.2020.103664 ↗
5. Gharghabi, S., Ding, Y., Yeh, C.C.M., Kamgar, K., Ulanova, L., and Keogh, E. (2017, January 18–21). Matrix profile VIII: Domain agnostic online semantic segmentation at superhuman performance levels. Proceedings of the 2017 IEEE International Conference on Data Mining (ICDM), New Orleans, LA, USA.10.1109/ICDM.2017.21 ↗
6. Domain agnostic online semantic segmentation for multi-dimensional time series10.1007/s10618-018-0589-3 ↗
7. SCHEDA: Lightweight Euclidean-like heuristics for anomaly detection in periodic time series10.1016/j.asoc.2019.105594 ↗
8. A novel computational approach for discord search with local recurrence rates in multivariate time series10.1016/j.ins.2018.10.047 ↗
9. An anomaly detection framework for time series data: An interval-based approach10.1016/j.knosys.2021.107153 ↗
10. Clustering-based anomaly detection in multivariate time series data10.1016/j.asoc.2020.106919 ↗
11. Swarms on the 3-sphere for online clustering of multivariate time series and data streams10.1016/j.future.2020.05.018 ↗
12. A fast LSH-based similarity search method for multivariate time series10.1016/j.ins.2018.10.026 ↗
13. Yeh, C.C.M., Zhu, Y., Ulanova, L., Begum, N., Ding, Y., Dau, H.A., Silva, D.F., Mueen, A., and Keogh, E. (2016, January 12–15). Matrix profile I: All pairs similarity joins for time series: A unifying view that includes motifs, discords and shapelets. Proceedings of the 2016 IEEE 16th International Conference on Data Mining (ICDM), Barcelona, Spain.10.1109/ICDM.2016.0179 ↗
14. Principal Component Analysis Based on Artificial Fish Swarm with T-Distribution Parameters10.1109/IAEAC50856.2021.9390973 ↗
15. (2022, March 10). Alexander Scarlat MD “Time Series with Anomalies”. Available online: https://www.kaggle.com/datasets/drscarlat/time-series.
16. Zhu, Y., Zimmerman, Z., Senobari, N.S., Yeh, C.C.M., Funning, G., Mueen, A., Brisk, P., and Keogh, E. (2016, January 12–15). Matrix profile II: Exploiting a novel algorithm and gpus to break the one hundred million barrier for time series motifs and joins. Proceedings of the 2016 IEEE 16th International Conference on data Mining (ICDM), Barcelona, Spain.10.1109/ICDM.2016.0085 ↗
17. Accelerating the similarity self-join using the GPU10.1016/j.jpdc.2019.06.005 ↗
18. Zhao, H., Wang, Y., Duan, J., Huang, C., Cao, D., Tong, Y., Xu, B., Bai, J., Tong, J., and Zhang, Q. (2020, January 17–20). Multivariate time-series anomaly detection via graph attention network. Proceedings of the 2020 IEEE International Conference on Data Mining, Sorrento, Italy.10.1109/ICDM50108.2020.00093 ↗
19. A spectral grouping and attention-driven residual dense network for hyperspectral image super-resolution10.1109/TGRS.2021.3049875 ↗
20. Goodness-of-fit testing for time series models via distance covariance10.1016/j.jeconom.2020.05.008 ↗
21. A solution to the dilemmalimiting similarity vs. limiting dissimilarity’by a method of transparent artificial intelligence10.1016/j.chaos.2021.110814 ↗
22. Diversity of deep earthquakes with waveform similarity10.1016/j.pepi.2021.106695 ↗
23. Rubinstein, B. (2020). A fast noise filtering algorithm for time series prediction using recurrent neural networks. arXiv.
24. Label consistent matrix factorization hashing for large-scale cross-modal similarity search10.1109/TPAMI.2018.2861000 ↗
25. Billion-scale similarity search with gpus10.1109/TBDATA.2019.2921572 ↗
26. Ma, L., Gu, X., and Wang, B. (2017). Correction of outliers in temperature time series based on sliding window prediction in meteorological sensor network. Information, 8.10.3390/info8020060 ↗
27. Dtwnet: A dynamic time warping network
28. Pelletier, C., Webb, G.I., and Petitjean, F. (2019). Temporal convolutional neural network for the classification of satellite image time series. Remote Sens., 11.10.3390/rs11050523 ↗
29. A hybrid method of exponential smoothing and recurrent neural networks for time series forecasting10.1016/j.ijforecast.2019.03.017 ↗
30. SSDTW: Shape segment dynamic time warping10.1016/j.eswa.2020.113291 ↗
31. Rubinstein, A., and Song, Z. (2020, January 5–8). Reducing approximate longest common subsequence to approximate edit distance. Proceedings of the Fourteenth Annual ACM-SIAM Symposium on Discrete Algorithms, Lake City, UT, USA.10.1137/1.9781611975994.98 ↗
32. An algorithm for outlier detection in a time series model using backpropagation neural network10.1016/j.jksus.2020.09.018 ↗
33. Linardi, M., Zhu, Y., Palpanas, T., and Keogh, E. (2018, January 10–15). Matrix profile X: VALMOD-scalable discovery of variable-length motifs in data series. Proceedings of the 2018 International Conference on Management of Data, Houston, TX, USA.10.1145/3183713.3183744 ↗
34. An interactive web-based dashboard to track COVID-19 in real time10.1016/S1473-3099(20)30120-1 ↗