Intrusion detection is very important now a days to ensure the safety of the computer networks. As, the cyber-attacks increasing these days traditional intrusion detection system which uses rule-based methods has failed to perform well. Machine learning assisted in providing solutions to these attacks and threats. Machine Learning observes patterns and learn from it to detect anomalies in the network traffic and help us to avoid those attacks. The intrusion detection using machine learning uses techniques like supervised, unsupervised, and deep learning to build an intrusion detection system that helps us to find out the attacks and threats in the network traffic. In this paper I have used a few machine learning algorithms like Decision-Tree, Random-Forest, Naïve-Bayes, Support-Vector-Machine, Logistic-Regression to classify the attacks in the dataset KDD CUP and test their accuracies to find out the train and test accuracies of the methods. Overall, the paper shows that the machine learning based intrusion detection system improves the network security.