Spanning Tree Protocol (STP) is an extensively used protocol in Ethernet networks that ensures loop-free paths by selectively blocking redundant links. However, STP is vulnerable to various attacks that can disrupt network operations, compromise network security, and lead to network downtime. This research proposes a novel approach to enhance the security and resilience of Ethernet networks by leveraging machine learning algorithms to detect and prevent attacks on the Spanning Tree Protocol. The core objective is to design a robust intrusion detection and prevention system that can adaptively safeguard the STP infrastructure in real-time. The first stage of the study involves collecting a comprehensive dataset of normal and attack scenarios in STP networks, including various attack types and their associated characteristics. Leveraging supervised learning techniques, machine learning models are trained on this dataset to learn the patterns and behaviors of different attacks on the STP. And finally, unsupervised learning algorithms are employed to analyze the network traffic and detect anomalies in real-time. This approach aims to identify any deviations from normal behavior, thereby indicating potential STP attacks.