English

Abstract

Listen

Translate article

Nowadays, Denial of service (DoS) attacks, have become a major security threat to networks and to the Internet, DoS is harmful to networks as it delays legitimate users from accessing the server, In general, some researches were done to detect and prevent DoS from occurring in a wide area network (WAN), but fewer researches were done on Local Area Network (LAN.), yet, detecting and preventing DoS attacks is still a challenging task, especially in LAN. In this paper, we propose an approach merging methods from data mining to detect and prevent DoS attacks, by using multi classification techniques to achieve a sufficient level of accuracy and reduce false alert alarm. And secondly, we will evaluate our approach in comparison with other existing approaches. Our work is based on EGH Dataset to detect DoS attacks, in addition, our approach is implemented using Rapidminer, the experimental results show that the proposed approach is effective in identifying DoS attacks, our designed approach achieves significant results. In the best case, our accuracy is up to 99.96%, we used two component of security; Snort tool and PfSense firewall, and compared our approach with other approaches, and we found that our approach achieves best accuracy results in most cases. Keywords— Data Mining, DoS attacks, intrusion detection, Misuse Detection, Multi Classification.