Millions of devices are connected by the Internet of Things (IoT), a quickly expanding field that enables little user participation in device interaction. Nevertheless, because the internet is public, IoT is vulnerable to several kinds of cyberattacks. To get around this, practical security measures like network detection algorithms are essential. Traditional security solutions are insufficient given the storage and computing power limits of IoT devices. Therefore, machine learning (ML)-based intelligent network-based solutions are essential. Although ML techniques for attack detection have been the subject of numerous studies in this paper, we evaluate different machine learning (ML) techniques for IoT network threat detection to close this gap. The authors of this study evaluate different machine learning (ML) techniques for IoT network alerting to close this gap. They apply six alternative machine learning algorithms and achieve high accuracy by using a recently generated dataset named Bot-IoT for evaluation. They also extract additional characteristics from the collected data that perform better than the features from earlier research. The study adds to the body of knowledge by showcasing successful ML-based strategies for IoT network security.