1 results listed
Intrusion detection systems (IDS) are critical to
assuring network security. These systems collect traffic data
from networks or systems and analyze it to identify potential
risks. Traditional methodologies, such as signature-based and
anomaly-based approaches, usually fail to adequately handle
the ever-changing nature of cyber threats. This work
investigates the use of machine learning approaches, notably
Random Forest and K-Nearest Neighbors (KNN), to improve
the detection capabilities of IDS. Random Forest uses many
decision trees to generate reliable classification results, whereas
KNN discovers anomalies by comparing them to established
patterns. The suggested approach demonstrated better
accuracy and precision in identifying intrusions after training
these models on recognized benchmark datasets and evaluated
their performance using key metrics. This study illustrates that
machine-learning-augmented IDS provides a comprehensive
and adaptive method to instant threat identification, solving the
limits of traditional techniques and advancing network security.
International Conference on Advanced Technologies, Computer Engineering and Science
ICATCES
Suhana Nafais A
Deja Chandru S
Harishkumar M
Sanjai B