Automating LIGO Glitch Witness Identification Using GravitySpy

Gravitational wave astronomy has made significant progress thanks to instruments like the Laser Interferometer Gravitational-wave Observatory (LIGO), which has greatly expanded our knowledge of colliding black holes and neutron stars. However, the high sensitivity of these detectors, along with the vast amount of data they generate, presents a challenge in distinguishing transient gravitational waves (GWs) from transient noise, or glitches, in the gravitational wave channel. Current methods include Omega Scan, a tool that provides detailed information on channels of interest surrounding a transient, and GravitySpy, a citizen-science project that uses human input to train a machine-learning (ML) model for glitch classification. We present OmegaNeuron, an innovative machine learning based tool that integrates both Omega Scan and GravitySpy to improve glitch witness identification and mitigation methods. OmegaNeuron has consistently identified relevant channels that may have witnessed these glitches, addressing the lack of effective glitch correlation tools.