Analysis of Machine Learning Techniques for Anomaly Detection in the Internet of Things

  • Shane Brady
  • , Damien Magoni
  • , John Murphy
  • , Haytham Assem
  • , A. Omar Portillo-Dominguez

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

A major challenge faced in the Internet of Things (IoT) is discovering issues that can occur in it, such as anomalies in the network or within the IoT devices. The nature of IoT hinders the identification of issues because of the huge number of devices and amounts of data generated. The aim of this paper is to investigate machine learning for effectively identifying anomalies in an IoT environment. We evaluated several state-of-the-art techniques which can identify, in real-time, when anomalies have occurred, allowing users to make alterations to the IoT network to eliminate the anomalies. Our results offer practitioners a valuable reference about which techniques might be more appropriate for their usage scenarios.

Original languageEnglish
Title of host publication2018 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538646250
DOIs
Publication statusPublished - 23 Jan 2019
Externally publishedYes
Event2018 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2018 - Gudalajara, Mexico
Duration: 6 Nov 20189 Nov 2018

Publication series

Name2018 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2018

Conference

Conference2018 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2018
Country/TerritoryMexico
CityGudalajara
Period6/11/189/11/18

Keywords

  • Anomaly Detection
  • Comparative Study
  • Internet of Things
  • Machine Learning

Fingerprint

Dive into the research topics of 'Analysis of Machine Learning Techniques for Anomaly Detection in the Internet of Things'. Together they form a unique fingerprint.

Cite this