Quantum machine learning for feature selection in Internet of Things network intrusion detection

Feature selection is a critical step in the machine learning (ML) model development workflow, aimed at identifying the most relevant subset of features from a dataset to improve ML model performance. In this paper, we investigate the use of quantum annealing to enhance the efficiency and effectiveness of feature selection, as compared to classical algorithmic methods for feature selection, prior to constructing a ML model for Internet of Things network intrusion detection. We aim to determine the optimal selection of network traffic features that contribute most to the detection of network intrusions. Leveraging a quantum annealing algorithm, which exploits quantum mechanics principles to find optimal solutions, along with D-Wave's hybrid quantum computing service, enables us to successfully tackle this combinatorial optimization problem. Our quantum machine learning approach leverages the strengths of both classical and quantum computing, offering a "one-shot" solution to feature selection without the need for iterative ML model training or incremental construction of the solution.