Two combination stages of clustered One-Class Classifiers for writer identification from text fragments
作者：
Highlights：
• Propose a new writer identification system based on text fragments.
• Use the clustered One-Class classifier for writer identification from text fragments.
• Propose a new dynamic fragment weighting combination rule.
• Propose the use of Choquet fuzzy integral to combine multiple systems.
• The two combination stages of the proposed system offers a competitive performance.
摘要
•Propose a new writer identification system based on text fragments.•Use the clustered One-Class classifier for writer identification from text fragments.•Propose a new dynamic fragment weighting combination rule.•Propose the use of Choquet fuzzy integral to combine multiple systems.•The two combination stages of the proposed system offers a competitive performance.
论文关键词：Clustered one-class classifiers,Dynamic fragment weighting combination,Choquet fuzzy integral,Writer identification,Text fragments
论文评审过程：Received 12 July 2017, Revised 22 April 2018, Accepted 3 May 2018, Available online 11 May 2018, Version of Record 15 June 2018.
论文官网地址：https://doi.org/10.1016/j.patcog.2018.05.001
