Fusing multi-scale fuzzy information to detect outliers

Baiyang Chen, Yongxiang Li, Dezhong Peng, Hongmei Chen, Zhong Yuan

Abstract: Highlights•A novel information fusion model based on multi-scale fuzzy granules is formulated.•An unsupervised outlier detection algorithm that integrates multi-scale information is proposed.•Extensive experiments indicate the effectiveness and advantages of our method over some state-of-the-arts.