The world is witnessing an unprecedented growth of needs in data analytics. Big Data is distinguished by its three main characteristics: velocity, variety and volume. An open issue and challenge faced by the data community is how to scale up analytic algorithms. To address this issue, optimization of large scale data sets has attracted many researchers in recent years. In this paper, we first present the most recent advances in optimization of Big Data analytics. Further, we introduce a fully distributed stochastic optimization algorithm for decision making over large scale data sets. We also propose the optimal weight design for the proposed algorithm and study its performance by considering a practical application in cognitive networks. Experimental results confirm that the proposed method performs well, proven to be distributed, scalable and robust to missing data and communication failures.

A new distributed and decentralized stochastic optimization algorithm with applications in Big Data analytics

Shahbazian R.;Grandinetti L.;Guerriero F.
2019

Abstract

The world is witnessing an unprecedented growth of needs in data analytics. Big Data is distinguished by its three main characteristics: velocity, variety and volume. An open issue and challenge faced by the data community is how to scale up analytic algorithms. To address this issue, optimization of large scale data sets has attracted many researchers in recent years. In this paper, we first present the most recent advances in optimization of Big Data analytics. Further, we introduce a fully distributed stochastic optimization algorithm for decision making over large scale data sets. We also propose the optimal weight design for the proposed algorithm and study its performance by considering a practical application in cognitive networks. Experimental results confirm that the proposed method performs well, proven to be distributed, scalable and robust to missing data and communication failures.
978-3-030-13708-3
978-3-030-13709-0
Big Data; Decision making; Distributed; Optimization; Stochastic
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/20.500.11770/302460
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 2
  • ???jsp.display-item.citation.isi??? ND
social impact