Intrusion Detection in Smart Homes Using K-Nearest Neighbors and Decision Trees Algorithm on IoT Network Traffic for Attack Classification

Abstract

Many homes now feature smart technology and numerous devices connected to the Internet, exposing them to cyberattacks. Therefore, implementing protection mechanisms to identify, predict, and mitigate these threats to smart home devices is crucial. This research proposes two machine learning models—K-Nearest Neighbors and Decision Tree—to predict malicious activity in smart home connections and classify whether an attack is occurring. The study presents both models along with an in-depth analysis of their performance, assessing how they function on unseen data and their effectiveness on the dataset. The findings highlight the strengths and weaknesses of each model, providing valuable insights into their applicability in real-world scenarios. By offering a comparative evaluation, this research contributes to the ongoing efforts in enhancing the security of smart homes and underscores the importance of adopting advanced machine learning techniques for intrusion detection systems (IDS). This study aims to lay the groundwork for future developments in smart home cybersecurity solutions.