In light of the exponential growth of the Internet of Things (IoT), the establishment of a robust IoT network security infrastructure has become an imperative. The IoT network generates dynamic traffic, and IoT devices continuously exchange data, necessitating precise categorization of IoT network traffic to optimise network performance, identify malicious activities, and detect anomalies. We have suggested machine learning approaches that categorise IoT traffic to address this issue. The categorization of traffic patterns can achieve efficient management of network resources and mitigate potential security threats. In this study, we presented a methodology for classifying IoT network traffic using machine learning techniques. We employ the XGBOOST classifier, Multilayer Perceptron (MLP), and 1-D Convolutional Neural Network (1DCNN) models to classify diverse IoT traffic categories based on their unique characteristics. Experimental results indicate that the XGBOOST model outperforms both 1D-CNN and MLP, with an accuracy of 99% as opposed to 98% for 1D-CNN and 96% for MLP.