Proceedings of the
The Nineteenth International Conference on Computational Intelligence and Security (CIS 2023)
December 1 – 4, 2023, Haikou, China
A Noise Points Detection Algorithm for Clustering
1School of Basic Medicine, Shaanxi University of Chinese Medicine, China.
2School of Computer Science, Xian Polytechnic University, China.
ABSTRACT
Noise points can reduce the performance of clustering algorithms and increase the time cost on clustering. In order to reduce the impact of noise points during the clustering process, a new noise points detection algorithm is proposed. The algorithm determines whether a data point is a noise point based on two indicators: the distance between data points and the number of natural neighbors of the data point. It not only considers the feature that the noise point is far away from other data points, but also considers the feature that the number of natural neighbors of the noise point is relatively small. Therefore, the algorithm is suitable for multiple distributed data sets. Experimental results show the efficacy of the proposed noise points detection algorithm.
Keywords: Noise detection, Clustering, Distance metric, Neighbor.
Download PDF
1School of Basic Medicine, Shaanxi University of Chinese Medicine, China.
2School of Computer Science, Xian Polytechnic University, China.
ABSTRACT
Noise points can reduce the performance of clustering algorithms and increase the time cost on clustering. In order to reduce the impact of noise points during the clustering process, a new noise points detection algorithm is proposed. The algorithm determines whether a data point is a noise point based on two indicators: the distance between data points and the number of natural neighbors of the data point. It not only considers the feature that the noise point is far away from other data points, but also considers the feature that the number of natural neighbors of the noise point is relatively small. Therefore, the algorithm is suitable for multiple distributed data sets. Experimental results show the efficacy of the proposed noise points detection algorithm.
Keywords: Noise detection, Clustering, Distance metric, Neighbor.
Download PDF