THE RESEARCH OF KNN AND SVM CLASSIFICATION PERFORMANCE ON TWO KINDS OF UNBALANCED DATA SET

DOI

10.2991/iccsee.2013.366How to use a DOI?

Keywords

Unbalanced Data Set, classify, KNN, SVM

Abstract

For Unbalanced Data Set, the KNN (K - the nearest neighbor) and SVM (support vector machine) classification algorithm’s prediction result would tend to most class; the misclassification rate of the minority class was big. This paper analyzed in detail the influence of unbalanced data set to KNN and SVM in theory, and proposed a new method to solve this problem. Experiment based on UCI data set using KNN and SVM algorithm to prove the validity of the proposed method.

Copyright

© 2013, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Juan DU
AU  - Li-li JIANG
PY  - 2013/03
DA  - 2013/03
TI  - THE RESEARCH OF KNN AND SVM CLASSIFICATION PERFORMANCE ON TWO KINDS OF UNBALANCED DATA SET
BT  - Proceedings of the 2nd International Conference on Computer Science and Electronics Engineering (ICCSEE 2013)
PB  - Atlantis Press
SP  - 1453
EP  - 1456
SN  - 1951-6851
UR  - https://doi.org/10.2991/iccsee.2013.366
DO  - 10.2991/iccsee.2013.366
ID  - DU2013/03
ER  -