Label disambiguation-based feature selection for partial label learning via fuzzy dependency and feature discernibility

Wenbin Qian, Jinfei Ding, Yihui Li, Jintao Huang

Abstract: Highlights•A novel instance distribution-based label disambiguation method is proposed.•A weighted fuzzy rough sets is presented to deal with fuzzy and uncertain information.•A label disambiguation-based feature selection is proposed for partial label learning.•Extensive experiments indicate that the proposed algorithm is effective and feasible.