Classification of gamma ray bursts using machine learning techniques

The classification of Gamma-Ray Bursts (GRBs) has been a subject of ongoing debate in the field of high-energy astrophysics. While the traditional bimodal classification based on T90 duration has been widely used, numerous studies have identified potential intermediate classes and classification anomalies that challenge this simplified scheme. Our methodology employs a comprehensive analysis of detailed temporal morphological features derived from Fermi-GBM light curves across multiple energy bands , capturing intrinsic shape characteristics including rise times, decay profiles etc. By applying unsupervised learning techniques to these morphological features, we aim to identify natural clustering patterns that may reveal previously unrecognized GRB classes. Our preliminary results suggest the presence of distinct morphological subgroups that do not strictly align with the conventional short-long dichotomy. This analysis could provide new insights into GRB progenitor populations and prompt a reassessment of current classification paradigms. We discuss the implications of our findings for understanding GRB physics and the potential need for a more nuanced classification scheme that better reflects the diversity of GRB phenomena.