A Hybrid Dimensionality Reduction for Network Intrusion Detection
Due to the wide variety of network services, many different types of protocols exist, producing various packet features. Some features contain irrelevant and redundant information. The presence of such features increases computational complexity and decreases accuracy. Therefore, this research is de...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-11-01
|
| Series: | Journal of Cybersecurity and Privacy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2624-800X/3/4/37 |
| _version_ | 1827574474654875648 |
|---|---|
| author | Humera Ghani Shahram Salekzamankhani Bal Virdee |
| author_facet | Humera Ghani Shahram Salekzamankhani Bal Virdee |
| author_sort | Humera Ghani |
| collection | DOAJ |
| description | Due to the wide variety of network services, many different types of protocols exist, producing various packet features. Some features contain irrelevant and redundant information. The presence of such features increases computational complexity and decreases accuracy. Therefore, this research is designed to reduce the data dimensionality and improve the classification accuracy in the UNSW-NB15 dataset. It proposes a hybrid dimensionality reduction system that does feature selection (FS) and feature extraction (FE). FS was performed using the Recursive Feature Elimination (RFE) technique, while FE was accomplished by transforming the features into principal components. This combined scheme reduced a total of 41 input features into 15 components. The proposed systems’ classification performance was determined using an ensemble of Support Vector Classifier (SVC), K-nearest Neighbor classifier (KNC), and Deep Neural Network classifier (DNN). The system was evaluated using accuracy, detection rate, false positive rate, f1-score, and area under the curve metrics. Comparing the voting ensemble results of the full feature set against the 15 principal components confirms that reduced and transformed features did not significantly decrease the classifier’s performance. We achieved 94.34% accuracy, a 93.92% detection rate, a 5.23% false positive rate, a 94.32% f1-score, and a 94.34% area under the curve when 15 components were input to the voting ensemble classifier. |
| first_indexed | 2024-03-08T20:38:44Z |
| format | Article |
| id | doaj.art-44349a3df4e24326b0ea0f1bd82fa1a5 |
| institution | Directory Open Access Journal |
| issn | 2624-800X |
| language | English |
| last_indexed | 2024-03-08T20:38:44Z |
| publishDate | 2023-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Cybersecurity and Privacy |
| spelling | doaj.art-44349a3df4e24326b0ea0f1bd82fa1a52023-12-22T14:17:45ZengMDPI AGJournal of Cybersecurity and Privacy2624-800X2023-11-013483084310.3390/jcp3040037A Hybrid Dimensionality Reduction for Network Intrusion DetectionHumera Ghani0Shahram Salekzamankhani1Bal Virdee2School of Computing and Digital Media, London Metropolitan University, London N7 8DB, UKSchool of Computing and Digital Media, London Metropolitan University, London N7 8DB, UKSchool of Computing and Digital Media, London Metropolitan University, London N7 8DB, UKDue to the wide variety of network services, many different types of protocols exist, producing various packet features. Some features contain irrelevant and redundant information. The presence of such features increases computational complexity and decreases accuracy. Therefore, this research is designed to reduce the data dimensionality and improve the classification accuracy in the UNSW-NB15 dataset. It proposes a hybrid dimensionality reduction system that does feature selection (FS) and feature extraction (FE). FS was performed using the Recursive Feature Elimination (RFE) technique, while FE was accomplished by transforming the features into principal components. This combined scheme reduced a total of 41 input features into 15 components. The proposed systems’ classification performance was determined using an ensemble of Support Vector Classifier (SVC), K-nearest Neighbor classifier (KNC), and Deep Neural Network classifier (DNN). The system was evaluated using accuracy, detection rate, false positive rate, f1-score, and area under the curve metrics. Comparing the voting ensemble results of the full feature set against the 15 principal components confirms that reduced and transformed features did not significantly decrease the classifier’s performance. We achieved 94.34% accuracy, a 93.92% detection rate, a 5.23% false positive rate, a 94.32% f1-score, and a 94.34% area under the curve when 15 components were input to the voting ensemble classifier.https://www.mdpi.com/2624-800X/3/4/37network securitynetwork traffic anomaliesintrusion detectiondimensionality reductionprincipal component analysisrecursive feature elimination |
| spellingShingle | Humera Ghani Shahram Salekzamankhani Bal Virdee A Hybrid Dimensionality Reduction for Network Intrusion Detection Journal of Cybersecurity and Privacy network security network traffic anomalies intrusion detection dimensionality reduction principal component analysis recursive feature elimination |
| title | A Hybrid Dimensionality Reduction for Network Intrusion Detection |
| title_full | A Hybrid Dimensionality Reduction for Network Intrusion Detection |
| title_fullStr | A Hybrid Dimensionality Reduction for Network Intrusion Detection |
| title_full_unstemmed | A Hybrid Dimensionality Reduction for Network Intrusion Detection |
| title_short | A Hybrid Dimensionality Reduction for Network Intrusion Detection |
| title_sort | hybrid dimensionality reduction for network intrusion detection |
| topic | network security network traffic anomalies intrusion detection dimensionality reduction principal component analysis recursive feature elimination |
| url | https://www.mdpi.com/2624-800X/3/4/37 |
| work_keys_str_mv | AT humeraghani ahybriddimensionalityreductionfornetworkintrusiondetection AT shahramsalekzamankhani ahybriddimensionalityreductionfornetworkintrusiondetection AT balvirdee ahybriddimensionalityreductionfornetworkintrusiondetection AT humeraghani hybriddimensionalityreductionfornetworkintrusiondetection AT shahramsalekzamankhani hybriddimensionalityreductionfornetworkintrusiondetection AT balvirdee hybriddimensionalityreductionfornetworkintrusiondetection |