Presentation #116.113 in the session Stellar/Compact Objects.
Commonly studied phenomena in heliophysics, such as solar flares, coronal mass ejections (CMEs), and sunspots, provide useful insights into both the universe and our own solar system. Recently, deep learning and artificial intelligence techniques have been harnessed to identify and understand correlations between various phenomena on the Sun’s outer atmospheres. To train supervised machine learning models, labeled data including bounding boxes and classification categories are often required. The lack of properly labeled data can hinder research progress in these nascent directions. Meanwhile, engaging citizen scientists in both astrophysics and machine learning research is an effective way to obtain cost-effective data and democratize science as a whole. In this work, we propose the deployment of HelioLable, a mobile application aimed at providing a platform for crowdsourcing ground-truth labels to identify solar phenomena.