Fast adaptively balanced min-cut clustering

Feiping Nie, Fangyuan Xie, Jingyu Wang, Xuelong Li

Abstract: Highlights•A fast adaptively balanced min-cut clustering is proposed to maximize intra-cluster similarity.•The balanced factors are added in to alleviate skewed clustering result.•A one-step optimization algorithm is proposed to solve the objective function.•Our proposed method could directly solve the discrete indicator matrix.•The experimental results on real datasets indicate the effectiveness of our proposed method.