Volume no :19, Issue no: 1, (2025)

International Journal of Applied Mathematics and Machine Learning

DATA CLASSIFICATION BASED ON K NEAREST NEIGHBOUR CLASSIFIER

Author's: Zhaoquan Cheng, Yuan Wang, Xiao Ding, Qirui Xiao, Zhiguo An, Zhengxin Guan and Jianqiang Gao
Pages: [13] - [20]
Received Date: August 3, 2025
Submitted by:
DOI: http://dx.doi.org/10.18642/ijamml_7100122325

Abstract

This paper implements the k-Nearest Neighbour (KNN) classifier on the Iris data set by using MATLAB software and systematically investigates how training set size and the neighbour parameter k influence classification accuracy. Experimental results indicate that when the training sample size reaches 120, the classification accuracy achieves 100%. The optimal performance (98.33%) is observed when k is in the range of 3 to 5. This study provides a basis for parameter optimization in classification tasks within data mining and offers practical guidance for decision support systems.

Keywords

KNN classifier, data mining, Iris data set, classification accuracy.