 This study proposes a machine learning based approach to detecting and mapping liquefaction ejecta from aerial imagery using advanced image processing techniques and compares its performance with supervised and semi-supervised learning methods. The proposed methodology considers available partial liquefaction labels as high certainty features and uses a combination of color transformation bands, statistical indices, texture components, and dimensionality reduction outputs to improve classification accuracy. Building footprints are also used to mask out building roofs from the classification process. The study shows that the semi-supervised method performs best when selected high-ranked features of statistical indices and dimensionality reduction outputs are used and can better augment the liquefaction labels across the image in terms of spatial completeness compared to supervised learning. This article was authored by Adela Sadi, Laurie Gaskins-Bays, Christina Sanon, and others.