SFKNN-DPC: Standard deviation weighted distance based density peak clustering algorithm

Juanying Xie, Xinglin Liu, Mingzhao Wang

Abstract: Highlights•Standard deviation weighted distance is proposed to enhance the Euclidean distance.•Local density ρi and distance δi of point i are defined utilizing the innovative distance, so do outliers and non-outliers.•Divide and conquer assignment strategy is proposed for assigning non-outliers and outliers in turn.•An innovative density peak clustering algorithm referred to SFKNN-DPC is proposed.•Extensive experiments demonstrate that SFKNN-DPC is superior to the peers in comparison.