An Improved kNN Algorithm Based on Conditional Probability Distance Metric

DOI

10.2991/icmmct-17.2017.211How to use a DOI?

Keywords

kNN, Nominal Variable, Distance Metric, Conditional Probability, QPSO

Abstract

There is no rational distance metric for nominal variables in traditional kNN classification algorithm. And the weighting methods commonly used in kNN cannot process datasets with multi-type variables or depend much on the field knowledge. An improved kNN method based on conditional probability and QPSO is presented is this paper. This approach measures the distance between two nominal variables by the distribution difference of instances' classes, which makes full use of attribute values' information. Meanwhile, it adopts QPSO to adjust attribute weight so that the weight will enhance the classification accuracy. This approach is able to process datasets with multi-type variables and less depends on parameters. Finally, experiments were taken on the UCI data set, which shows that our approach is superior in performance to algorithms compared.

Copyright

© 2017, 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  - Ziyang Liu
AU  - Zhanbao Gao
AU  - Xulong Li
PY  - 2017/04
DA  - 2017/04
TI  - An Improved kNN Algorithm Based on Conditional Probability Distance Metric
BT  - Proceedings of the 2017 5th International Conference on Machinery, Materials and Computing Technology (ICMMCT 2017)
PB  - Atlantis Press
SP  - 1057
EP  - 1062
SN  - 2352-5401
UR  - https://doi.org/10.2991/icmmct-17.2017.211
DO  - 10.2991/icmmct-17.2017.211
ID  - Liu2017/04
ER  -