Quantum-Inspired Clustering for Hazardous Asteroid Prediction in Quantum Machine Learning

Abstract

An asteroid impact is one of the rare natural disasters that can be prevented or mitigated using the proper preparation and preparatory measures. The main goal is to investigate the use of quantum machine learning (QML) in the context of asteroid prediction in order to improve early detection and trajectory forecasting capabilities. New computational approaches are necessary in the dynamic field of astronomical hazard assessment, and QML offers itself as an advanced paradigm to meet the challenges of this important task. In this study, we evaluate the EQIE-FCM (Enhanced Quantum-Inspired Evolutionary Fuzzy C-Means) clustering algorithm and compare it with other models such as K-Medoid, Spectral Clustering, Fuzzy C-Means, Quantum K-Means, and Quantum Fuzzy C-Means. EQIE-FCM outperforms these models, surpassing Silhouette and Davies-Bouldin thresholds. The choice of clustering algorithm depends on data characteristics and problem context. By leveraging quantum computing to evolve crucial parameters, EQIE-FCM effectively clusters datasets. We evaluate its efficacy using different-sized asteroid datasets. Quantum machine learning shows promise for accurate predictions of hazardous asteroids, but its integration requires awareness of both strengths and limitations.