Title :
An improved KNN method and its application to tumor diagnosis
Author :
Yuan, Wei ; Liu, Juan ; Zhou, Huai-Bei
Author_Institution :
Sch. of Comput., Wuhan Univ., China
Abstract :
In this paper, we present an improved KNN (k-nearest neighbor) classification method called KNN+. In order to estimate its performance, we compare it with the original KNN method in the area of gene expression based tumor diagnosis, by the aid of two gene selection methods. The experiment results show that KNN+ is superior to KNN in general.
Keywords :
feature extraction; genetics; patient diagnosis; pattern classification; support vector machines; tumours; gene expression based tumor diagnosis; gene selection method; improved KNN classification method; k-nearest neighbor classification method; Cancer; Euclidean distance; Gene expression; Machine learning; Neoplasms; Noise reduction; Performance analysis; Support vector machine classification; Support vector machines; Testing;
Conference_Titel :
Machine Learning and Cybernetics, 2004. Proceedings of 2004 International Conference on
Print_ISBN :
0-7803-8403-2
DOI :
10.1109/ICMLC.2004.1378515
