2017/09/24 by Michael P. B. Gallaugher, Paul D. McNicholas, Gallaugher, Michael P. B. +1
Computer Science · #Bayesian Methods and Mixture Models #Computation (stat.CO) #FOS: Computer and information sciences #Methodology (stat.ME)
paper · pdf · doi:10.48550/arxiv.1709.08258
openalex publication_date 2017/09/24 · openalex created_date 2022/10/05 · openalex updated_date 2026/07/28
Recent work on fractionally-supervised classification (FSC), an approach that\nallows classification to be carried out with a fractional amount of weight\ngiven to the unlabelled points, is further developed in two respects. The\nprimary development addresses a question of fundamental importance over how to\nchoose the amount of weight given to the unlabelled points. The resolution of\nthis matter is essential because it makes FSC more readily applicable to real\nproblems. Interestingly, the resolution of the weight selection problem opens\nup the possibility of a different approach to model selection in model-based\nclustering and classification. A secondary development demonstrates that the\nFSC approach can be effective beyond Gaussian mixture models. To this end, an\nFSC approach is illustrated using mixtures of multivariate t-distributions.\n