R. Qahwaji, M. Al-Omari, T. Colak, and S. Ipson (UK)
Space Weather, Solar Imaging, Machine learning, RBF, Prediction, and ROC.
In this paper, computer association algorithms are designed to study the associations between solar features (eruptive filaments/prominences) and the extremes of space weather (Coronal Mass Ejections (CMEs)). Years of solar data are analysed in this study to identify patterns of associations that can be represented using computerised learning rules to enable real-time and reliable CME predictions. RBF networks are used for our machine learning experiments. The NGDC filaments catalogue and the SOHO/LASCO CMEs catalogue are processed to associate filaments with CMEs based on timing and location information. Features representing the filament time, duration, type and extent are extracted from all the associated (A) and not-associated (NA) filaments and converted to a numerical format that is suitable for RBF learning and testing. The machine learning system predicts if the filament is likely to initiate a CME. Intensive experiments are carried out to determine the group of input features that could provide the best prediction performance.
Important Links:
Go Back
