The Internet of Things (IoT) ecosystem consists of interconnected devices which are continuously growing and billions of devices are continuously sharing data over a variety of network protocols. However, these protocols are often vulnerable to cyber-attacks, making security a critical concern. In order to identify and report attacks, intrusion detection systems, or IDS have become a viable method for identifying and thwarting malicious activities in Internet of Things environments. In this work, we create an ensemble model that can differentiate between malicious and legitimate network traffic in an effort to make intrusion detection systems (IDS) more effective. This ensemble-based method combines a variety of machine learning techniques to improve model robustness, decrease overfitting, and improve classification accuracy. We used CSE-CIC-IDS2018 dataset to train our ensemble model, which comprised GaussianNB, Multi-Layer Perceptron, and Random Forest as the basic classifier. Using voting ensemble learning strategy, we integrated these models to further improve detection accuracy. In cybersecurity applications, our testing results demonstrated superiority over the suggested methods for managing threat detection and identifying anomalous components.