The usage of the Internet of Things (IoT) in today’s world has led to several stern security issues like denialof-service attacks by a huge collection of compromised IoT devices. Due to lack of proper security and unavailability of packet filtration, the IoT devices are easily compromised and can be a member of the zombie network. In spite of addressing several techniques in detecting IoT botnets, unaddressed challenges are still open to researchers. In this paper, a few machine-learning methods are introduced to recognize the existence of IoT botnets effectively. The machine-learning model detects the prediction of the IoT botnets based up on the information on the network traffic. Our proposed model achieves less false positive for faster detection by using feature selection. The Random Forest came up with an accuracy of 94.47 percent, which performed much better than other deep learning and machine learning models and, thus, can be measured as a suitable explanation to effectually sense the IoT botnet with a lesser detection rate.