TY - GEN

T1 - An incremental nearest neighbor algorithm with queries

AU - Ratsaby, Joel

PY - 1998/1/1

Y1 - 1998/1/1

N2 - We consider the general problem of learning multi-category classification from labeled examples. We present experimental results for a nearest neighbor algorithm which actively selects samples from different pattern classes according to a querying rule instead of the a priori class probabilities. The amount of improvement of this query-based approach over the passive batch approach depends on the complexity of the Bayes rule. The principle on which this algorithm is based is general enough to be used in any learning algorithm which permits a model-selection criterion and for which the error rate of the classifier is calculable in terms of the complexity of the model.

AB - We consider the general problem of learning multi-category classification from labeled examples. We present experimental results for a nearest neighbor algorithm which actively selects samples from different pattern classes according to a querying rule instead of the a priori class probabilities. The amount of improvement of this query-based approach over the passive batch approach depends on the complexity of the Bayes rule. The principle on which this algorithm is based is general enough to be used in any learning algorithm which permits a model-selection criterion and for which the error rate of the classifier is calculable in terms of the complexity of the model.

UR - http://www.scopus.com/inward/record.url?scp=84898991074&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84898991074

SN - 0262100762

SN - 9780262100762

T3 - Advances in Neural Information Processing Systems

SP - 612

EP - 618

BT - Advances in Neural Information Processing Systems 10 - Proceedings of the 1997 Conference, NIPS 1997

PB - Neural information processing systems foundation

T2 - 11th Annual Conference on Neural Information Processing Systems, NIPS 1997

Y2 - 1 December 1997 through 6 December 1997

ER -