A hybrid approach for intrusion detection based on machine learning
by Rohit Singh; Mala Kalra; Shano Solanki
International Journal of Security and Networks (IJSN), Vol. 15, No. 4, 2020

Abstract: With the evolution of internet, network security has emerged as one of the key areas of research. Network security is to safeguard the privacy, availability and integrity of the system. Identification of intrusion is core component to attain overall network security success. Therefore, intrusion detection (ID) is broadly explored by researchers and idea of identification of intrusion has developed into a system, known as intrusion detection system (IDS). IDS focuses on investigation of attacks and offers desirable support for defence management along with information about the intrusion. Several intrusion detection approaches are already proposed to mitigate the impact of intrusion. In this paper, these techniques are discussed to highlight the state-of-the-art and a multilevel hybrid approach based on SVM-Naïve is proposed to predict malicious traffic from the network. Here, KNN along with k-fold cross validation is used to create new high-quality training dataset that outstandingly improved performance of classifiers. Performance of proposed model is evaluated by using standard KDD 1999 Cup dataset. Comparison of proposed model is carried out with existing techniques and it is analysed that its performance is better regarding exactness (99.73%), recognition rate of attack classes (99.81%) and FPR (0.21%).

Online publication date: Tue, 10-Nov-2020

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Security and Networks (IJSN):
Login with your Inderscience username and password:

    Username:        Password:         

Forgotten your password?


Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com