International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 10 Issue: 04 | Apr 2023
p-ISSN: 2395-0072
www.irjet.net
Application of neural network and PSO-SVM in intrusion detection of network Gopika S1, Samitha T2 1Dept. of Electronics and Communication Engineering, Mahaguru Institute of Technology, Kerala 2Asst. Professor, Dept. of Electronics and Communication Engineering, Mahaguru Institute of Technology, Kerala
---------------------------------------------------------------------***--------------------------------------------------------------------publicly accessible NSL-KDD network dataset. It includes Abstract – Imbalanced network traffic can often be a
data on network traffic with 41 traffic features. A new deeplearning approach for intrusion detection based on the NSLKDD dataset is presented in this paper. Deep learning and machine learning are the basis of the proposed effort. It applies the Support Vector Machine (SVM) classification algorithm, Convolutional Neural Network (CNN) feature extraction technique, and Particle Swarm Optimization (PSO) SVM algorithm optimization. In Chapter 3, the system description is explained. The experimental result of the system is presented in Chapter 4. The conclusion of the work is given in Chapter 5.
gateway for malicious cyber-attacks to penetrate networks and go undetected. In these situations, it is challenging for Network Intrusion Detection System (NIDS) to find the attacker since they can blend in with a lot of normal data. An intrusion detection system (IDS) monitors network traffic for suspicious activities and immediately provides notifications if it detects anything suspicious. The IDS looks for any activity that might be a sign of an attack or intrusion by comparing the network activity to a set of predetermined rules and patterns. Even the most sophisticated NIDS may have trouble identifying this type of assault because of its high degree of stealth and obfuscation in cyberspace. A new approach based on deep learning and machine learning using NSL-KDD dataset for intrusion detection is proposed in this paper. The proposed approach uses an SVM classifier for the attack classification task and a 1-Dimensional Convolutional Neural Network for feature extraction.
2. LITERATURE REVIEW A network intrusion system based on Naive Bayes has been suggested in [4]. Across data sets that have been tagged by the services, the framework develops the network service patterns. The naive Bayes Classifier method, together with the built-in patterns, allows the framework to identify attacks in the datasets. This approach has a greater detection rate, requires less time to complete, and is less expensive than the neural network-based approach. However, it produces more false positives than true ones.
Key Words:
Machine learning, Deep learning, Convolutional Neural Network (CNN), Support Vector Machine (SVM), Particle Swarm Optimization (PSO)
1. INTRODUCTION
When it comes to meeting the demands of contemporary networks, there are questions about the viability and sustainability of current systems. These worries are more directly related to the declining levels of detection accuracy and the rising levels of required human intervention. To address these concerns, a deep learning-based NIDS approach was proposed in [5]. This unique deep-learning classification model was developed using stacked NDAEs.
Cybersecurity faces tremendous risks as a result of the rapid advancement of technologies like 5G, IoT, cloud computing, and others that have increased network scale, real-time traffic, and cyberattack complexity and diversity [1][2]. Security breaches might sneak in with a lot of regular traffic. As a result, it is simple to misclassify because the machine learning algorithm cannot fully learn the distribution of some categories. Most of the newly generated cyber-attacks are created by subtly altering already known ones, which is typically handled as regular traffic on the IoT network [3].
In order to address the issue of network traffic domain model architecture design, a network architecture search algorithm (NAS) in the field of network traffic together with a surrogate model have been suggested in [6]. Under the premise of a specified optimization target, a neural architecture search (NAS) can automatically search the model's architecture. A surrogate model was used in the network architecture search task to determine how candidate architectures would perform. This approach increases the effectiveness of the architecture search and, to a certain extent, solves the issues of the network search algorithm's need for large computing resources and significant time consumption.
To find unusual or hostile activity in the network, a system called Network Intrusion Detection System (NIDS) is utilized. IDS keeps an eye out for harmful activity in network traffic. There are numerous ways to identify suspicious activity in network communications. IDS monitors network traffic persistently to look for network intrusions. A recent trend in many security applications is to combine deep learning methodologies with cybersecurity because of their excellent performance. For analysis, the system needs a dataset with past traffic data. The most widely utilized dataset is the
© 2023, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 1287