• Nishant J Seth
• Trish Bandhekar
• Sonu Yadav

As technology has been evolving at a rapid speed, it has led to more vulnerable data and that has increased the number of unauthorized access attempts. To overcome this, an Intrusion Detection System is used. An IDS is used to detect any such malicious attempts. Current generation of IDS’ cannot detect complex attacks and take way too long to detect when using high dimensional data. Other problems of IDS include the high false alarm and low detection. To solve these drawbacks, we propose a system that uses machine learning based technique to identify these malicious packets in a low amount of time. We use a technique known as Principal Component Analysis (PCA) to reduce our high dimensional dataset into a lower dimensional dataset while still keeping our accuracy up without too much loss of data. Random Forest is used as a classification algorithm to detect our packets. An accuracy of 0.996 has been obtained. This experiment was conducted on the UNSW-NB15 dataset.

Intrusion Detection System (IDS), Principal Component Analysis (PCA), Random Forest, Classification

"Intrusion Detection System using PCA and Random Forest", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 4, page no.776-779, April-2019, Available :http://www.jetir.org/papers/JETIR1904G19.pdf

"Intrusion Detection System using PCA and Random Forest", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 4, page no. pp776-779, April-2019, Available at : http://www.jetir.org/papers/JETIR1904G19.pdf