Organizations are becoming increasingly vulnerable to potential cyber threats which defined as network intrusions. Intrusion Detection is basically providing the security or managing the flow of data, information, managing the access of the system to only authorized user. In adaptive false alarms filter a combination of machine learning algorithms is used to increase the classification accuracy of the system. The experimental results in this research show that the proposed J48, Decision Table and k-star techniques reducing the false alarm rate and improving the accuracy. The new NSL-KDD dataset is used, which is applied with WEKA tool.