A2095
Title: Predicting X-ray plateau occurrence in gamma-ray bursts using statistical and machine learning approaches
Authors: Nuha Al Maashari - Sultan Qaboos University (Oman) [presenting]
Iman Al Hasani - Sultan Qaboos University (Oman)
Abstract: The occurrence of an X-ray plateau phase in gamma-ray burst (GRB) afterglows is investigated to determine whether it can be predicted from observable burst properties. Data from the Swift/XRT catalogue, including spectral, energetic, temporal, and redshift-related properties, were analysed using multiple logistic regression, Random Forest classification, dimensionality reduction, and clustering methods. Both supervised models showed limited predictive performance, with Random Forest providing improved discrimination compared with logistic regression but without achieving reliable classification. Dimensionality reduction and clustering analyses further revealed substantial overlap between plateau and non-plateau GRBs, indicating that the observed properties do not form distinct groups associated with plateau behaviour. These findings suggest that X-ray plateau formation is governed by complex multivariate processes that are not fully represented by the available observable characteristics. The findings highlight the limitations of current GRB observables for plateau prediction and demonstrate the value of combining statistical modelling and machine learning approaches in astrophysical data analysis.
