Joint exploring of risky labeled and unlabeled samples for safe semi-supervised clustering
作者：
Highlights：
• We propose a novel approach to jointly explore labeled and unlabeled samples.
• Local density and minimum distance are designed to measure the safe degree.
• A local graph is constructed to safely exploit the risky prior knowledge.
摘要
•We propose a novel approach to jointly explore labeled and unlabeled samples.•Local density and minimum distance are designed to measure the safe degree.•A local graph is constructed to safely exploit the risky prior knowledge.
论文关键词：Safe semi-supervised clustering,Fuzzy c-means,Safe degree,Local density,Minimum distance
论文评审过程：Received 28 August 2019, Revised 9 February 2021, Accepted 25 February 2021, Available online 10 March 2021, Version of Record 30 March 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.114796
