Predictive Classification and Bayesian Inference
Jie Xiong · Työväentutkimus Vuosikirja · 2015
A general inductive probabilistic framework for clustering and classi-fication is introduced using the principles of Bayesian predictive in-ference, such that all quantities are jointly modelled and the uncer-tainty is fully acknowledged through the posterior predictive distri-bution. Several learning rules have been considered and the theoreti-cal results are extended to acknowledge complex dependencies within the datasets. Multiple probabilistic models have been developed for analysing data from a wide variate of fields of applications. State-of-art algorithms are introduced and developed for the model optimiza-tion. iii iv Acknowledgements I am grateful for the funding provided by the Finnish Doctoral Programme