Signature-based anomaly intrusion detection using integrated data mining classifiers

As the influence of Internet and networking technologies as communication medium advance and expand across the globe, cyber attacks also grow accordingly. Anomaly detection systems (ADSs) are employed to scrutinize information such as packet behaviours coming from various locations on network to fin...

Full description

Bibliographic Details
Main Authors: Yassin, Warusia, Udzir, Nur Izura, Abdullah, Azizol, Abdullah @ Selimun, Mohd Taufik, Zulzalil, Hazura, Muda, Zaiton
Format: Conference or Workshop Item
Language:English
Published: IEEE 2014
Online Access:http://psasir.upm.edu.my/id/eprint/47759/
http://psasir.upm.edu.my/id/eprint/47759/1/Signature-based%20anomaly%20intrusion%20detection%20using%20integrated%20data%20mining%20classifiers.pdf
_version_ 1848850898418139136
author Yassin, Warusia
Udzir, Nur Izura
Abdullah, Azizol
Abdullah @ Selimun, Mohd Taufik
Zulzalil, Hazura
Muda, Zaiton
author_facet Yassin, Warusia
Udzir, Nur Izura
Abdullah, Azizol
Abdullah @ Selimun, Mohd Taufik
Zulzalil, Hazura
Muda, Zaiton
author_sort Yassin, Warusia
building UPM Institutional Repository
collection Online Access
description As the influence of Internet and networking technologies as communication medium advance and expand across the globe, cyber attacks also grow accordingly. Anomaly detection systems (ADSs) are employed to scrutinize information such as packet behaviours coming from various locations on network to find those intrusive activities as fast as possible with precision. Unfortunately, besides minimizing false alarms; the performance issues related to heavy computational process has become drawbacks to be resolved in this kind of detection systems. In this work, a novel Signature-Based Anomaly Detection Scheme (SADS) which could be applied to scrutinize packet headers' behaviour patterns more precisely and promptly is proposed. Integrating data mining classifiers such as Naive Bayes and Random Forest can be utilized to decrease false alarms as well as generate signatures based on detection results for future prediction and reducing processing time. Results from a number of experiments using DARPA 1999 and ISCX 2012 benchmark dataset have validated that SADS own better detection capabilities with lower processing duration as contrast to conventional anomaly-based detection method.
first_indexed 2025-11-15T10:13:36Z
format Conference or Workshop Item
id upm-47759
institution Universiti Putra Malaysia
institution_category Local University
language English
last_indexed 2025-11-15T10:13:36Z
publishDate 2014
publisher IEEE
recordtype eprints
repository_type Digital Repository
spelling upm-477592016-07-15T03:42:46Z http://psasir.upm.edu.my/id/eprint/47759/ Signature-based anomaly intrusion detection using integrated data mining classifiers Yassin, Warusia Udzir, Nur Izura Abdullah, Azizol Abdullah @ Selimun, Mohd Taufik Zulzalil, Hazura Muda, Zaiton As the influence of Internet and networking technologies as communication medium advance and expand across the globe, cyber attacks also grow accordingly. Anomaly detection systems (ADSs) are employed to scrutinize information such as packet behaviours coming from various locations on network to find those intrusive activities as fast as possible with precision. Unfortunately, besides minimizing false alarms; the performance issues related to heavy computational process has become drawbacks to be resolved in this kind of detection systems. In this work, a novel Signature-Based Anomaly Detection Scheme (SADS) which could be applied to scrutinize packet headers' behaviour patterns more precisely and promptly is proposed. Integrating data mining classifiers such as Naive Bayes and Random Forest can be utilized to decrease false alarms as well as generate signatures based on detection results for future prediction and reducing processing time. Results from a number of experiments using DARPA 1999 and ISCX 2012 benchmark dataset have validated that SADS own better detection capabilities with lower processing duration as contrast to conventional anomaly-based detection method. IEEE 2014 Conference or Workshop Item PeerReviewed application/pdf en http://psasir.upm.edu.my/id/eprint/47759/1/Signature-based%20anomaly%20intrusion%20detection%20using%20integrated%20data%20mining%20classifiers.pdf Yassin, Warusia and Udzir, Nur Izura and Abdullah, Azizol and Abdullah @ Selimun, Mohd Taufik and Zulzalil, Hazura and Muda, Zaiton (2014) Signature-based anomaly intrusion detection using integrated data mining classifiers. In: International Symposium on Biometrics and Security Technologies (ISBAST 2014), 26-27 Aug. 2014, Kuala Lumpur, Malaysia. (pp. 232-237). 10.1109/ISBAST.2014.7013127
spellingShingle Yassin, Warusia
Udzir, Nur Izura
Abdullah, Azizol
Abdullah @ Selimun, Mohd Taufik
Zulzalil, Hazura
Muda, Zaiton
Signature-based anomaly intrusion detection using integrated data mining classifiers
title Signature-based anomaly intrusion detection using integrated data mining classifiers
title_full Signature-based anomaly intrusion detection using integrated data mining classifiers
title_fullStr Signature-based anomaly intrusion detection using integrated data mining classifiers
title_full_unstemmed Signature-based anomaly intrusion detection using integrated data mining classifiers
title_short Signature-based anomaly intrusion detection using integrated data mining classifiers
title_sort signature-based anomaly intrusion detection using integrated data mining classifiers
url http://psasir.upm.edu.my/id/eprint/47759/
http://psasir.upm.edu.my/id/eprint/47759/
http://psasir.upm.edu.my/id/eprint/47759/1/Signature-based%20anomaly%20intrusion%20detection%20using%20integrated%20data%20mining%20classifiers.pdf