• Megha Tomar
• Avinash Pal
• Trapti Ozha
• Durgesh Mishra

On the internet, harmful behaviours that harm a single machine or the entire network may occur. They may be random. Computer connection is growing at an unprecedented rate, making it tougher to stay up. The internet may pose security risks, just as it might in person. IDS software monitors a network for suspicious or malicious behaviour. IDS is a technology that helps detect system assaults and identify their perpetrators. In the past, machine learning (ML) approaches were employed to improve IDS accuracy and intruder detection outcomes. In this post, we discuss how to build an IDS using PCA and NN classification method. This is a strategy for building a good IDS. PCA may be used to minimise the dimensionality of the data and organise it, whereas random forest can categorise it. The suggested procedure will be applied on the KDD (Knowledge Discovery Dataset). It's evident that the offered approach is more accurate than SVM, Naive Bayes, and Decision Tree. One can be confident of this because: Our technique yielded the following results: Performance time is 2.25 minutes, accuracy is 98.35%, and error rate is 0.12.

IDS, Knowledge Discovery Dataset, PCA, Random Forest.

"Intrusion Detection System using PCA with NN Approach", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 8, page no.a463-a467, August-2022, Available :http://www.jetir.org/papers/JETIR2208061.pdf

"Intrusion Detection System using PCA with NN Approach", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 8, page no. ppa463-a467, August-2022, Available at : http://www.jetir.org/papers/JETIR2208061.pdf