• Likith Gowda S
• Rini Sharma
• Dr. Jagruthi H

The frequency of cyberattacks has increased due to the Internet's rapid proliferation. Intrusion detection systems (IDS) are being used to protect system security. IDS is still experiencing some issues making its categorization function better. First off, the complexity of high-dimensional qualities presents a barrier to the effectiveness and speed of the categorization for IDS. Second, the classification performance of the0traditional Stacking approach is directly influenced by the basic classifiers. To solve the two difficulties described above, we offer a0hybrid intrusion detection system based on a weighted Stacking classification method and a CFS-DE feature selection strategy. To minimize the dimension of the features, we employed the CFS-DE algorithm, which looks for the ideal feature subset. Then, a weighted Stacking technique is recommended to improve classification performance by decreasing the weights of the fundamental classifiers with bad training results and increasing the weights of those with positive outcomes. The model thus enhances classification efficiency and accuracy. For all experiments in0this work, the NSL-KDD and CSE-CIC-IDS2018 data sets were utilized.

Intrusion Detection System(IDS), Deep Learning, Machine Learning.

"ANOMALY DETECTION IN NETWORK TRAFFIC USING DEEP LEARNING TECHNIQUES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 5, page no.i601-i607, May-2023, Available :http://www.jetir.org/papers/JETIR2305903.pdf

"ANOMALY DETECTION IN NETWORK TRAFFIC USING DEEP LEARNING TECHNIQUES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 5, page no. ppi601-i607, May-2023, Available at : http://www.jetir.org/papers/JETIR2305903.pdf