Surrounding Influenced K-Nearest Neighbors:A New Distance Based Classifier

摘要：

The nearest neighbor classification method assigns to an unclassified point the class of the nearest of a set of previously classified points. An extension to this approach is the if-NN method, in which the classification is made taking into account the K nearest points and classifying the unclassified point by a voting criteria from this k points. We present a new method that extends the K-NN limits, taking into account, for each neighbor, its / nearest neighbors. Experimental results are promising, obtaining better results for two class problems than the original KNN.

关键词： Nearest Neighbor Supervised Classification

作者: I. Mendialdua B. Sierra E. Lazkano I. Irigoien E. Jauregi

作者单位: Department of Computer Science and Artificial Intelligence University of the Basque Country P. Manuel Lardizabal 120018 Donostia-San Sebastian Basque Country Spain

会议类型: 国际会议

会议名称: 6th International Conference on Advanced Data Mining and Applications(第六届先进数据挖掘及应用国际会议 ADMA 2010)

会议地点: 重庆

会议语种:英文

页码: 270-277

在线出版日期: 2010-11-19（万方平台首次上网日期，不代表论文的发表时间）