We address classification problems for which the training instances are governed by a distribution that is allowed to differ arbitrarily from the test distribution--problems also ...
In this paper, we propose a probabilistic kernel approach to preference learning based on Gaussian processes. A new likelihood function is proposed to capture the preference relat...
Abstract. Use of document genre in information retrieval systems has the potential to improve the task-appropriateness of results. However, genre classification remains a challengi...
Luanne Freund, Charles L. A. Clarke, Elaine G. Tom...
This study emphasizes the importance of using appropriate measures in particular text classification settings. We focus on methods that evaluate how well a classifier performs. The...
We propose a generalization of multilabel classification that we refer to as graded multilabel classification. The key idea is that, instead of requesting a yes-no answer to the q...