Parallel Streaming Implementation of Online Time Series Correlation Discovery on Sliding Windows with Regression Capabilities

Boyan Kolev (1) , Reza Akbarinia (1) , Ricardo Jimenez-Peris (2) , Oleksandra Levchenko (1) , Florent Masseglia (1) , Marta Patino (3) , Patrick Valduriez (1)

1 ZENITH - Scientific Data Management
2 DSL - Distributed Systems Laboratory
3 UPM - Universidad Politécnica de Madrid

This paper addresses the problem of continuously finding highly correlated pairs of time series over the most recent time window and possibly use the discovered correlations to select features for training a regression model for prediction. The implementation builds upon the ParCorr parallel method for online correlation discovery and is designed to run continuously on top of the UPM-CEP data streaming engine through efficient streaming operators.

Mots clés: Distributed Computing; Time Series Correlation and Regression; Data Stream Processing

Domaines: Recherche d'information [cs.IR]

Boyan Kolev, Reza Akbarinia, Ricardo Jimenez-Peris, Oleksandra Levchenko, Florent Masseglia, et al.. Parallel Streaming Implementation of Online Time Series Correlation Discovery on Sliding Windows with Regression Capabilities. CLOSER 2019 - 9th International Conference on Cloud Computing and Services Science, May 2019, Heraklion, Greece. pp.681-687, ⟨10.5220/0007843806810687⟩. ⟨lirmm-02265729⟩