A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things

Shengchu Zhao, Wei Li, Tanveer Zia, Albert Y. Zomaya

Research output: Book chapter/Published conference paperConference paper

5 Citations (Scopus)

Abstract

Internet of Things (IoT) devices and services have gained wide spread growth in many commercial and mission critical applications. The devices and services suffer from intrusions, attacks and malicious activities. To protect valuable data transmitted through IoT networks and users’ privacy, intrusion detection systems (IDS) should be developed to match with the characteristics of IoT, which requires real-time monitoring. This paper proposes a novel model for intrusion detection which is based on dimension reduction algorithm and a classifier, which can be used as an online machine learning algorithm. The proposed model uses Principal Component Analysis (PCA) to reduce dimensions of dataset from a large number of features to a small number. To develop a classifier, softmax regression and k-nearest neighbour algorithms are applied and compared. Experimental results using KDD Cup 99 Data Set show that our proposed model performs optimally in labelling benign behaviours and identifying malicious behaviours. The computing complexity and time performance approve that the model can be used to detect intrusions in IoT.
Original languageEnglish
Title of host publicationA Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things
Subtitle of host publication15th IEEE International Conference on Pervasive Intelligence and Computing (PICom 2017)
Place of PublicationUnited States
Pages836-843
Number of pages8
ISBN (Electronic) 9781538619551
DOIs
Publication statusPublished - 02 Apr 2018
Event15th IEEE International Conference on Pervasive Intelligence and Computing: PICOM 2017 - Holiday Inn Disney Spring, Orlando, United States
Duration: 06 Nov 201710 Nov 2017
http://cse.stfx.ca/~picom2017/ (Conference website)
https://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=8327304 (Conference proceedings)

Conference

Conference15th IEEE International Conference on Pervasive Intelligence and Computing
CountryUnited States
CityOrlando
Period06/11/1710/11/17
Internet address

Fingerprint

Intrusion detection
Classifiers
Principal component analysis
Labeling
Learning algorithms
Learning systems
Internet of things
Monitoring

Cite this

Zhao, S., Li, W., Zia, T., & Zomaya, A. Y. (2018). A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things. In A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things: 15th IEEE International Conference on Pervasive Intelligence and Computing (PICom 2017) (pp. 836-843). United States. https://doi.org/10.1109/DASC-PICom-DataCom-CyberSciTec.2017.141
Zhao, Shengchu ; Li, Wei ; Zia, Tanveer ; Zomaya, Albert Y. / A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things. A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things: 15th IEEE International Conference on Pervasive Intelligence and Computing (PICom 2017). United States, 2018. pp. 836-843
@inproceedings{87b0fae102f44ad3864e17641358ec6b,
title = "A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things",
abstract = "Internet of Things (IoT) devices and services have gained wide spread growth in many commercial and mission critical applications. The devices and services suffer from intrusions, attacks and malicious activities. To protect valuable data transmitted through IoT networks and users’ privacy, intrusion detection systems (IDS) should be developed to match with the characteristics of IoT, which requires real-time monitoring. This paper proposes a novel model for intrusion detection which is based on dimension reduction algorithm and a classifier, which can be used as an online machine learning algorithm. The proposed model uses Principal Component Analysis (PCA) to reduce dimensions of dataset from a large number of features to a small number. To develop a classifier, softmax regression and k-nearest neighbour algorithms are applied and compared. Experimental results using KDD Cup 99 Data Set show that our proposed model performs optimally in labelling benign behaviours and identifying malicious behaviours. The computing complexity and time performance approve that the model can be used to detect intrusions in IoT.",
author = "Shengchu Zhao and Wei Li and Tanveer Zia and Zomaya, {Albert Y.}",
year = "2018",
month = "4",
day = "2",
doi = "10.1109/DASC-PICom-DataCom-CyberSciTec.2017.141",
language = "English",
pages = "836--843",
booktitle = "A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things",

}

Zhao, S, Li, W, Zia, T & Zomaya, AY 2018, A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things. in A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things: 15th IEEE International Conference on Pervasive Intelligence and Computing (PICom 2017). United States, pp. 836-843, 15th IEEE International Conference on Pervasive Intelligence and Computing, Orlando, United States, 06/11/17. https://doi.org/10.1109/DASC-PICom-DataCom-CyberSciTec.2017.141

A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things. / Zhao, Shengchu; Li, Wei; Zia, Tanveer; Zomaya, Albert Y.

A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things: 15th IEEE International Conference on Pervasive Intelligence and Computing (PICom 2017). United States, 2018. p. 836-843.

Research output: Book chapter/Published conference paperConference paper

TY - GEN

T1 - A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things

AU - Zhao, Shengchu

AU - Li, Wei

AU - Zia, Tanveer

AU - Zomaya, Albert Y.

PY - 2018/4/2

Y1 - 2018/4/2

N2 - Internet of Things (IoT) devices and services have gained wide spread growth in many commercial and mission critical applications. The devices and services suffer from intrusions, attacks and malicious activities. To protect valuable data transmitted through IoT networks and users’ privacy, intrusion detection systems (IDS) should be developed to match with the characteristics of IoT, which requires real-time monitoring. This paper proposes a novel model for intrusion detection which is based on dimension reduction algorithm and a classifier, which can be used as an online machine learning algorithm. The proposed model uses Principal Component Analysis (PCA) to reduce dimensions of dataset from a large number of features to a small number. To develop a classifier, softmax regression and k-nearest neighbour algorithms are applied and compared. Experimental results using KDD Cup 99 Data Set show that our proposed model performs optimally in labelling benign behaviours and identifying malicious behaviours. The computing complexity and time performance approve that the model can be used to detect intrusions in IoT.

AB - Internet of Things (IoT) devices and services have gained wide spread growth in many commercial and mission critical applications. The devices and services suffer from intrusions, attacks and malicious activities. To protect valuable data transmitted through IoT networks and users’ privacy, intrusion detection systems (IDS) should be developed to match with the characteristics of IoT, which requires real-time monitoring. This paper proposes a novel model for intrusion detection which is based on dimension reduction algorithm and a classifier, which can be used as an online machine learning algorithm. The proposed model uses Principal Component Analysis (PCA) to reduce dimensions of dataset from a large number of features to a small number. To develop a classifier, softmax regression and k-nearest neighbour algorithms are applied and compared. Experimental results using KDD Cup 99 Data Set show that our proposed model performs optimally in labelling benign behaviours and identifying malicious behaviours. The computing complexity and time performance approve that the model can be used to detect intrusions in IoT.

UR - http://cse.stfx.ca/~picom2017/ps.php

UR - http://cse.stfx.ca/~picom2017/2017%20Advanced%20Program%2018Oct_final.pdf

UR - http://cse.stfx.ca/~picom2017/index.php

U2 - 10.1109/DASC-PICom-DataCom-CyberSciTec.2017.141

DO - 10.1109/DASC-PICom-DataCom-CyberSciTec.2017.141

M3 - Conference paper

SP - 836

EP - 843

BT - A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things

CY - United States

ER -

Zhao S, Li W, Zia T, Zomaya AY. A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things. In A Dimension Reduction Model and Classifier for Anomaly-Based Intrusion Detection in Internet of Things: 15th IEEE International Conference on Pervasive Intelligence and Computing (PICom 2017). United States. 2018. p. 836-843 https://doi.org/10.1109/DASC-PICom-DataCom-CyberSciTec.2017.141