Selective Sampling for Nearest Neighbor Classi ers

In the passive, traditional, approach to learning, the information available to the learner is a set of classiied examples, which are randomly drawn from the instance space. In many applications, however, the initial clas-siication of the training set is a costly process, and an intelligently selection of training examples from unla-beled data is done by an active learner. This paper proposes a lookahead algorithm for example selection and addresses the problem of active learning in the context of nearest neighbor classiiers. The proposed approach relies on using a random eld model for the example labeling, which implies a dynamic change of the label estimates during the sampling process. The proposed selective sampling algorithm was evaluated empirically on artiicial and real data sets. The experiments show that the proposed method outper-forms other methods in most cases.