A Proof of Concept in Multivariate Time Series Clustering Using Recurrent Neural Networks and SP-Lines

Big Data and the IoT explosion has made clustering multivariate Time Series (TS) one of the most effervescent research fields. From Bio-informatics to Business and Management, multivariate TS are becoming more and more interesting as they allow to match events the co-occur in time but that is hardly noticeable. This study represents a step forward in our research. We firstly made use of Recurrent Neural Networks and transfer learning to analyze each example, measuring similarities between variables. All the results are finally aggregated to create an adjacency matrix that allows extracting the groups. In this second approach, splines are introduced to smooth the TS before modeling; also, this step avoid to learn from data with high variation or with noise. In the experiments, the two solutions are compared suing the same proof-of-concept experimentation.

[1]  Hector Budman,et al.  A hybrid clustering approach for multivariate time series - A case study applied to failure analysis in a gas turbine. , 2017, ISA transactions.

[2]  Svetlana Simic,et al.  A Preliminary Study on Multivariate Time Series Clustering , 2019, SOCO.

[3]  Ke Yan,et al.  A novel computational approach for discord search with local recurrence rates in multivariate time series , 2019, Inf. Sci..

[4]  Witold Pedrycz,et al.  Multivariate time series anomaly detection: A framework of Hidden Markov Models , 2017, Appl. Soft Comput..

[5]  Ying Wah Teh,et al.  Time-series clustering - A decade review , 2015, Inf. Syst..

[6]  Guangyu Liu,et al.  Time series clustering and physical implication for photovoltaic array systems with unknown working conditions , 2019, Solar Energy.

[7]  Carlos Arthur Mattos Teixeira Cavalcante,et al.  Pattern recognition as a tool to support decision making in the management of the electric sector. Part II: A new method based on clustering of multivariate time series , 2015 .

[8]  Pierpaolo D'Urso,et al.  Robust fuzzy clustering of multivariate time trajectories , 2018, Int. J. Approx. Reason..

[9]  Dirk Müller,et al.  A time series clustering approach for Building Automation and Control Systems , 2019 .

[10]  Qiang Yang,et al.  A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.

[11]  Leanne Lai Hang Chan,et al.  A fast LSH-based similarity search method for multivariate time series , 2019, Inf. Sci..

[12]  Witold Pedrycz,et al.  Time-series clustering based on linear fuzzy information granules , 2018, Appl. Soft Comput..

[13]  Giuseppe Nunnari,et al.  Multivariate time series clustering on geophysical data recorded at Mt. Etna from 1996 to 2003 , 2013 .

[14]  Robert Jenssen,et al.  Time Series Cluster Kernel for Learning Similarities between Multivariate Time Series with Missing Data , 2017, Pattern Recognit..

[15]  Won Bo Lee,et al.  Robust design of ambient-air vaporizer based on time-series clustering , 2018, Comput. Chem. Eng..

[16]  Stephen Grossberg,et al.  Recurrent neural networks , 2013, Scholarpedia.