Cyber security in smart home Internet of Things devices: Threat detection and prevention using artificial intelligence

The Internet of Things (IoT) has permeated every element of life in this day and age of technology, including smart environments, smart homes and smart scenarios. Many IoT devices in smart homes operate non-stop and in big quantities. Living in such areas might be more serene with improved security and authentication of these smart gadgets. To ensure that smart IoT devices operate flawlessly, it is crucial to keep an eye on their actions. These devices are readily attacked by hackers because of their tiny size, low power and resource consumption and ease of use. It is essential to defend the smart home environment’s features and integrity against outside threats. Machine learning (ML) has been essential in recognising these kinds of harmful efforts and actions. There are several ML techniques available to identify both typical and anomalous IoT device traffic behaviour. This research suggested an anomaly detection method for smart homes based on ML and several classifiers. The BoT-IoT dataset from the University of New South Wales (UNSW) is used for testing and assessment. Using a dataset of IoT devices, ML models based on four classifiers are constructed. Random forest, decision tree and AdaBoost have weighted precision, recall and F1 score of 1 for the test dataset, but an artificial neural network (ANN) has 0.98, 0.96 and 0.96, accordingly. This article is also included in The Business & Management Collection which can be accessed at https://hstalks.com/business/.