Relief-based feature selection: Introduction and review

• Published in: Journal of biomedical informatics Vol. 85; pp. 189 - 203
• Main Authors: Urbanowicz, Ryan J., Meeker, Melissa, La Cava, William, Olson, Randal S., Moore, Jason H.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.09.2018
• Subjects: Algorithms; Computational Biology - methods; Data Mining - methods; Epistasis; Feature interaction; Feature selection; Feature weighting; Filter; Humans; Models, Statistical; Regression Analysis; ReliefF; Software; Feature selection; Epistasis; Filter; Feature interaction; ReliefF; Feature weighting

[Display omitted] •Relief-based feature selection methods (RBAs) are reviewed in detailed context.•RBAs can detect interactions without examining pairwise combinations.•Iterative RBAs have been developed to scale them up to very large feature spaces.•Research focused on core algorithms, iterative sc...