Machine learning algorithms for spatio-temporal data mining
Ranga Raju Vatsavai · University of Minnesota Digital Conservancy (University of Minnesota) · 2008
Remote sensing, which provides inexpensive, synoptic-scale data with multi-temporal coverage, has proven to be very useful in land cover mapping, environmental monitor-ing, forest and crop inventory, urban studies, natural and man made object recognition, etc. Thematic information extracted from remote sensing imagery is also useful in vari-ety of spatiotemporal applications. However, increasing spatial, spectral, and temporal resolutions invalidate several assumptions made by the traditional classification meth-ods. In this thesis we addressed four specific problems, namely, small training samples, multisource data, aggregate classes, and spatial autocorrelation. We developed a novel semi-supervised learning algorithm to address the small training sample problem. A common assumption made in previous works is that the labeled and unlabeled training samples are drawn from the same mixture model. However, in practice we observed that the number of mixture components for labeled and unlabeled training samples differ significantly. Our adaptive semi-supervised algorithm over comes this impor-tant limitation by eliminating unlabeled samples from additional components through