Anomaly detection has become a crucial
technology in several application fields, mostly for network
security. The classification challenge of anomaly detection
using machine learning techniques on network data has
been described here. Using the KDD99 dataset for network
IDS, dimensionality reduction and classification
techniques are investigated and assessed. For the
application on network data, Principal Component
Analysis for dimensionality reduction and Support Vector
Machine for classification have been taken into
consideration, and the results are examined.. The result
shows the decrease in execution time for the classification
as we reduce the dimension of the input data and also the
precision and recall parameter values of the classification
algorithm shows that the SVM with PCA method is more
accurate as the number of misclassification decreases.
Enormous data in health research is extremely interesting
since data-based studies may move more quickly than
hypothesis-based research, despite the fact that enormous
databases are becoming common and hence challenging to
interpret. Using Principal Component Analysis (PCA), one
may make some datasets less dimensional. enhances
interpretability while retaining most of the information. It
does this by introducing fresh variables that are unrelated
to one another.
Keywords :
Machine Learning, Principal Component Analysis, Dimensionality Reduction, Intrusion Detection, Anomaly Detection, Principal Component Analysis, Support Vector Machine.