In this paper, we propose a dual distance-weighted voting for KNN, which can solve the oversmoothing of increasing the neighborhood size k. The proposed classifier is compared with the other methods on ten UCI data sets. Experimental results suggest that the proposed classifier is a promising algorithm due to its satisfactory classification performance and robustness over a large value of k. © 2011 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Gou, J., Luo, M., & Xiong, T. (2011). Improving k-nearest neighbor rule with dual weighted voting for pattern classification. In Communications in Computer and Information Science (Vol. 159 CCIS, pp. 118–123). https://doi.org/10.1007/978-3-642-22691-5_21
Mendeley helps you to discover research relevant for your work.