| Analysis of Intrusion Detection Using Machine Learning Techniques |
| Pages: 84-92 (10) | [Full Text] PDF (921 KB) |
| P NERLIKAR, S PANDEY, S SHARMA, S BAGADE |
|
Usha Mittal Institute of Technology, Department Of Computer Science and Technology, S.N.D.T Womens University Mumbai, 2020 https://doi.org/10.47277/IJCNCS/8(10)1 |
|
Abstract - Whenever an intrusion occurs into the computer system, the protection and computer assets are being compromised. Network-based attacks make it difficult for legitimate users to access various network services by purposely occupying or sabotaging network resources and services. The mechanisms such as, sending of large amounts of network traffic, exploiting well-known faults in networking services, and by overloading network hosts, the services are made unavailable. Hence, there is a desire for an Intrusion Detection System (IDS). Detection and analysis of such IDS is challenging. In this paper, we aim at detecting and analysing computer based attacks by examining various data records, for example KDDCup99 dataset. In this paper, the KDDCup99 dataset is employed for analyzing intrusion using machine learning techniques, like Support Vector Machine (SVM). For the efficiency, the dataset is reduced using data pre-processing methods i.e irrelevant or redundant features are removed. As a result, the accuracy and performance is evaluated using the Support Vector Machine Classification Algorithm. The accuracy obtained by our mechanism is 96.5%. The performance is evaluated by the error rate, so in our proposed system, we got an error rate as 3.38% which is observed to be efficient in terms of performance. |
| Index Terms - Accuracy, Intrusion Detection, Machine Learning, Network traffic, SVM |
C itation - P NERLIKAR, S PANDEY, S SHARMA, S BAGADE. "Analysis of Intrusion Detection Using Machine Learning Techniques." International Journal of Computer Networks and Communications Security 8, no. 10 (2020): 84-92. |
| A Deep Learning based approach for DDoS attack detection in IoT-enabled smart environments |
| Pages: 93-99 (7) | [Full Text] PDF (609 KB) |
| UM Badamasi, S Khaliq, O Babalola, S Musa, T Iqbal |
|
Changchun University of Science and Technology China Government College University Faisalabad Pakistan Northwestern Polytechnical University Xian China Northeastern University Shenyang, Liaoning Province China https://doi.org/10.47277/IJCNCS/8(10)2 |
|
Abstract - This research contributes to enhance the security of smart environments such as Internet-of-Things (IoT) network. IoT provides a big network to connect things around the world in concern to reduce human effort and to make a digitalized world more easily controllable. The heterogeneous and vulnerable nature of IoT network makes its environment a big target for cyber hackers. The ever-increasing nature of cyber threats as well as rapid development and growth of IoT infrastructure generates a severe need for the security of smart networks. One of the major threats that can take down the complete availability of targeted system is DDoS (Distributed Denial of Service) attack, which recently has attacked numerous IoT networks lead to enormous losses. Therefore, in order to secure IoT networks, the authors propose Deep Learning (DL) based Cuda-enabled Long-Short-Term-Memory (LSTM) technique with evaluation using latest CICDDoS2019 dataset for detection of DDoS attacks. The proposed system achieves highest accuracy of 99.60% and proves itself best as compared to current state-of-the-art solutions in the domain. Finally, we cross validate our results for the indication of unbiased performance. |
| Index Terms - IoT, DDoS, Deep Learning, LSTM, CICDDoS2019 |
C itation - UM Badamasi, S Khaliq, O Babalola, S Musa, T Iqbal. "A Deep Learning based approach for DDoS attack detection in IoT-enabled smart environments." International Journal of Computer Networks and Communications Security 8, no. 10 (2020): 93-99. |