Learning from partially supervised data using mixture models and belief functions

Learning from partially supervised data using mixture models and belief functions

This paper addresses classification problems in which the class membership of training data is only partially known. Each learning sample is assumed to consist in a feature vector xi ∈ X and an imprecise and/or uncertain “soft” label mi defined as a Dempster-Shafer basic belief assignment over the set of classes. This framework thus generalizes many kinds of learning problems including supervis...

Semi-Supervised Learning of Mixture Models

This paper analyzes the performance of semisupervised learning of mixture models. We show that unlabeled data can lead to an increase in classification error even in situations where additional labeled data would decrease classification error. We present a mathematical analysis of this “degradation” phenomenon and show that it is due to the fact that bias may be adversely affected by unlabeled ...

Boosting Mixture Models for Semi-supervised Learning

Partially Specified Belief Functions

This paper presents a procedure to determine a complete belief function from the known values of belief for some of the subsets of the frame of discerment. The method is based on the principle of minimum commitment and a new principle called the focusing principle. This additional principle is based on the idea that belief is specified for the most relevant sets: the focal elements. The resulti...

Induction of decision trees from partially classified data using belief functions

The work described in this report concerns the problem of reasoning with uncertainty, in particular the problem of building classifiers based on uncertain data. The kind of uncertainty we have been concerned with has been uncertain classification, i.e., classification of data for which the classification labels are not crisp. We have proposed a method to handle this kind of uncertainty that is ...