This work tackles the problem of learning a robust classification function from a very small sample set when a related but unlabeled data set is provided. To this end we define a new semi-supervised method that is based on a stability criterion. We successfully apply our proposal in the specific case of automatic diagnosis of intestinal motility disease using video capsule endoscopy.

We propose a semi-supervised framework to handle diverse data formats or data with mixedtype attributes. Our preliminary results in clustering data with mixed numerical and categorical attributes show that the proposed semi-supervised framework gives better clustering results in the categorical domain. Thus the seeds obtained from clustering the numerical domain give an additional knowledge to ...

AbstrAct The insufficiency of labeled training samples is a major obstacle in automatic semantic analysis of large

We investigate omni-supervised learning, a special regime of semi-supervised learning in which the learner exploits all available labeled data plus internet-scale sources of unlabeled data. Omni-supervised learning is lowerbounded by performance on existing labeled datasets, offering the potential to surpass state-of-the-art fully supervised methods. To exploit the omni-supervised setting, we p...

There has been a surge of interest in learning using a mix of labeled and unlabeled data. General approaches include semi-supervised learning and tranductive inference. In this paper we look at some of the unique ways in which unlabeled data can improve performance when doing link-based classification, the classification of objects making use of both object descriptions and the links between ob...

In this paper we investigate the effect of unlabeled data on generative classifiers in semi-supervised learning. We first characterize situations where unlabeled data cannot change estimates obtained with labeled data, and argue that such situations are unusual in practice. We then report on a large set of experiments involving labeled and unlabeled data, and demonstrate that unlabeled data can...

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 ...

The focus of this thesis is on learning approaches for what we call “low-quality data” and in particular data in which only small amounts of labeled target data is available. The first part provides background discussion on low-quality data issues, followed by preliminary study in this area. The remainder of the thesis focuses on a particular scenario: multi-view semi-supervised learning. Multi...

Partial cognates are pairs of words in two languages that have the same meaning in some, but not all contexts. Detecting the actual meaning of a partial cognate in context can be useful for Machine Translation tools and for Computer-Assisted Language Learning tools. In this paper we propose a supervised and a semisupervised method to disambiguate partial cognates between two languages: French a...