In the digital world, online base network communication increased significantly. Cause of this, numerous cyberattacks occur every day and create a challenge for the network system to identify that intrusion on time. In addition, within the network, we cannot ignore the existence of intruders as they can launch many harmful cyberattacks. An intrusion detection system is the most valuable technique to help prevent network systems by investigating the network traffic. There are many intrusions detection research has been conducted. However, intrusion detection system still has some research gap and challenges that need to be improved to detect the accuracy of new intrusions. For this purpose, current artificial intelligence-based approaches, especially machine learning (ML) and deep learning (DL), are highly capable of intrusion detection in a network system. Both methods provide accurate outcomes. We proposed a novel, fast, efficient intrusion detection technique based on an ensemble classifier. The suggested classifier integrates all votes of the classifiers from the various instances that operate hard voting to reach the absolute voted class identity label. However, it has been shown on standard datasets of Network Security Laboratory and Knowledge Discovery in Databases (KDD). The suggested method acquired the most heightened precision of 98.19%.