Utilizing intrusion detection systems is necessary to safeguard information systems from attacker attacks. Many publicly accessible open-source assault datasets have been released in recent years so that academics and also researchers can evaluate the performance of different detection classifiers. These datasets contain a full collection of exemplary network features. This study, researchers look at the problem of Network-Based Intrusion-Detection System (NIDS) by employing the Bot-IoT dataset from the network based Internet of Things (IoT) to evaluate its usefulness of seven distinct Ensemble Learning Classifiers in terms of detection efficiency (ELCs). The outcomes of our trial demonstrated that CatBoost was the ELC that performed the most effectively with Effectiveness, Positive predictive value, F-Measure, Training and Test Time, despite the fact that all ELCs had excellent classification metric scores.