Quantum Artificial Hummingbird algorithm for Feature Selection of Social IoT

The Internet of Things (IoT) benefits from social networking platforms in establishing and enhancing social-oriented services, information, autonomous relationships. Social IoT (SIoT) systems can boost the user experience real world several applications, including healthcare, transportation, entertainment. However, collected data various interconnected SIoT is massive, demanding robust efficient processing algorithms, feature extraction, selection, inference. This work presents an enhanced Artificial Hummingbird algorithm (AHA) for selection (FS). version AHA performed using advantages Quantum-based optimization. main aim Quantum to improve population’s exploration ability while discovering feasible regions. Extensive experiments utilizing eighteen UCI datasets were conducted validate developed FS method, QAHA. QAHA compared with other methods, experimental established its efficiency. Moreover, a set four are used evaluate applicability real-world setting. results these indicate high performance increase accuracy by decreasing number features. In case datasets, average 93% among datasets. Whereas, has nearly 90.7%, 98.7%, 92.2%, 84.6% Trajectory, GAS sensors, Hepatitis, MovementAAL respectively.