核心概念
다중 뷰 학습에서 갈등이 있는 데이터를 신뢰할 수 있는 방식으로 처리하는 중요성
統計資料
"Experiments performed on 6 datasets verify the effectiveness of ECML." (6개 데이터셋에서 수행된 실험은 ECML의 효과를 검증함.)
引述
"Most previous works assume that multiple views are strictly aligned."
"Real-world applications usually require making decisions for conflictive instances rather than only eliminating them."
"ECML first learns view-specific evidence, which could be termed as the amount of support to each category collected from data."