CNC-NOS: Class noise cleaning by ensemble filtering and noise scoring

• Published in: Knowledge-based systems Vol. 140; pp. 27 - 49
• Main Authors: Luengo, Julián, Shim, Seong-O, Alshomrani, Saleh, Altalhi, Abdulrahman, Herrera, Francisco
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 15.01.2018; Elsevier Science Ltd
• Subjects: Class noise; Classification; Data preprocessing; Data relabeling; Data reparation; Filtering software; Filtration; Noise; Noise filtering; Noisy data; Numerical controls; Data relabeling; Data preprocessing; Classification; Class noise; Noise filtering; Noisy data; Data reparation

•A novel class noise cleaner able not only to remove noisy instances but to successfully relabel them.•It introduces an improvement of the noise score measure already proposed in a recent filter, making it sensitive to the proximity of the neighbors.•A wide range of noise levels are used to compare...