click here to download project abstract
–
ABSTRACT
Intrusion Detection System (IDS) needs a data and for this it is important to keep the real
working environment to find out all the possibilities of how an attack is about to happen and this
seem to be expensive. A Software to detect network attacks in a computer network from
unidentified users, including known personnel. The attack detector‟s learning task works up a
predictive model which is a classifier in this case which differentiates the “bad” (i.e., intrusions
or attacks) and “good” or “normal” connections. The primary aim is to use machine learning
based techniques to provide packet connection transfer in a better way by predicting results with
the at most accuracy. Comparing and discussing the outputs from the couple of machine learning
algorithms used for the given dataset with evaluated classification report, find the confusion
matrix and categorize the data from priority and the result which shows that the efficiency of the
claimed machine learning algorithm method is to be compared with the best accuracy techniques
such as Precision, Recall and F1 Score.
