Portable Intrusion Detection in IoT Enabled Smart City Networks: A Comparative Study of Machine Learning Models

The Internet of Things (IoT) is transforming manyindustries through the integration of the Internet with electronicand mechanical devices, as well as sensors, to make more intelligentsystems in applications such as healthcare, smart homes,and public safety. This rapid growth in IoT devices has resultedin significant security risks, turning these systems into targets forcybercriminals. To counter these weaknesses, Network IntrusionDetection Systems are deployed, which monitor network trafficto identify potential malicious activities. This paper evaluates theperformance of machine learning models in detecting intrusionswithin IoT-enabled smart city networks. For this, the UNSWNB15dataset is used, which contains realistic network traffic. Thedataset was preprocessed, including handling missing data, onehotencoding categorical variables, and normalizing numericalfeatures. The paper performs multi-class classification to identifyspecific attack types. We tested various machine learningalgorithms, including Decision Tree, K-Nearest Neighbor, LinearRegression, etc. classifiers. The preprocessed dataset contained61 attributes with 81,173 entries, which was sufficient for themodels to be thoroughly tested. The results provide a lot ofinsight into the strengths and weaknesses of various machinelearning techniques in improving the security of IoT networks,especially in critical applications in smart cities.Index Terms—internet of things, machine learning, securityrisks, smart cities