Portelli, K. and Anagnostopoulos, C. (2017) Leveraging Edge Computing through Collaborative Machine Learning. In: 2017 5th International Conference on Future Internet of Things and Cloud Workshops (FiCloudW), Prague, Czech Republic, 21-23 Aug 2017, pp. 164-169. ISBN 9781538632819 (doi: 10.1109/FiCloudW.2017.72)
|
Text
142024.pdf - Accepted Version 594kB |
Abstract
The Internet of Things (IoT) offers the ability to analyze and predict our surroundings through sensor networks at the network edge. To facilitate this predictive functionality, Edge Computing (EC) applications are developed by considering: power consumption, network lifetime and quality of context inference. Humongous contextual data from sensors provide data scientists better knowledge extraction, albeit coming at the expense of holistic data transfer that threatens the network feasibility and lifetime. To cope with this, collaborative machine learning is applied to EC devices to (i) extract the statistical relationships and (ii) construct regression (predictive) models to maximize communication efficiency. In this paper, we propose a learning methodology that improves the prediction accuracy by quantizing the input space and leveraging the local knowledge of the EC devices.
Item Type: | Conference Proceedings |
---|---|
Status: | Published |
Refereed: | Yes |
Glasgow Author(s) Enlighten ID: | Portelli, Mr Kurt and Anagnostopoulos, Dr Christos |
Authors: | Portelli, K., and Anagnostopoulos, C. |
College/School: | College of Science and Engineering > School of Computing Science |
ISBN: | 9781538632819 |
Published Online: | 20 November 2017 |
Copyright Holders: | Copyright © 2017 IEEE |
First Published: | First published in 2017 5th International Conference on Future Internet of Things and Cloud Workshops (FiCloudW): 146-169 |
Publisher Policy: | Reproduced in accordance with the publisher copyright policy |
University Staff: Request a correction | Enlighten Editors: Update this record