The ubiquity of Internet of Things (IoT) gadgets in smart homes has transformed our interactions with our living environments by providing never-before-seen levels of automation and convenience. However, because IoT devices are becoming possible targets for malicious attacks, this broad connectivity also poses serious security risks. Ensuring the privacy, safety, and integrity of smart home ecosystems requires prompt detection and mitigation of these threats. Data from IoT devices is gathered, pre-processed, feature engineered, labelled, and divided into training, validation, and testing sets as part of a machine learning method to threat detection in smart home IoT networks. The process of choosing and training appropriate machine learning models—which can include everything from classification techniques to anomaly detection algorithms—is crucial. Methods are surveyed to review different types of cyber-attacks, such as denial-of-service (DoS), distributed denial-of-service (DDoS), probing, user-to-root (U2R), remote-to-local (R2L), botnet attack, spoofing, and man-in-the-middle (MITM) attacks. To protect user information, data anonymization and encryption techniques are used with privacy considerations. Another strategy that has been put forth aims to improve the security of IoT networks in smart homes by providing a strong defence against new threats and equipping users with the information and resources they need to keep their connected world safe. To provide a full overview of the numerous advancements in this field, a list of all works published in the literature to date is incorporated. Lastly, the study also includes suggestions for future research directions.