Distributed Algorithms to Find Similar Time Series

Oleksandra Levchenko (1) , Boyan Kolev (1) , Djamel-Edine Edine Yagoubi (1) , Dennis Shasha (2) , Themis Palpanas (3) , Patrick Valduriez (1) , Reza Akbarinia (1) , Florent Masseglia (1)

1 ZENITH - Scientific Data Management
2 NYU - New York University [New York]
3 UPD5 - Université Paris Descartes - Paris 5

As sensors improve in both bandwidth and quantity over time, the need for high performance sensor fusion increases. This requires both better (quasi-linear time if possible) algorithms and paral-lelism. This demonstration uses financial and seismic data to show how two state-of-the-art algorithms construct indexes and answer similarity queries using Spark. Demo visitors will be able to choose query time series, see how each algorithm approximates nearest neighbors and compare times in a parallel environment.

Mots clés: Distributed data processing; Spark; Similarity search; Indexing; Time series

Domaines: Recherche d'information [cs.IR]

Oleksandra Levchenko, Boyan Kolev, Djamel-Edine Edine Yagoubi, Dennis Shasha, Themis Palpanas, et al.. Distributed Algorithms to Find Similar Time Series. ECML-PKDD 2019 - European Conference on Machine Learning and Knowledge Discovery in Databases, Sep 2019, Wurtzbourg, Germany. pp.781-785, ⟨10.1007/978-3-030-46133-1_51⟩. ⟨lirmm-02265726⟩