Scalable multi-output label prediction: From classifier chains to classifier trellises
作者：
Highlights：
• A study of multi-output classification as graphical models.
• An empirical comparison of existing strategies for modelling dependency among outputs.
• A novel scalable approach based on a hill climbing heuristic: the classifier trellis.
• An empirical cross-fold comparison with other methods.
• A connection to structured output prediction and a comparison in a segmentation task.
摘要
Highlights•A study of multi-output classification as graphical models.•An empirical comparison of existing strategies for modelling dependency among outputs.•A novel scalable approach based on a hill climbing heuristic: the classifier trellis.•An empirical cross-fold comparison with other methods.•A connection to structured output prediction and a comparison in a segmentation task.
论文关键词：Classifier chains,Multi-label classification,Multi-output prediction,Structured inference,Bayesian networks
论文评审过程：Received 10 June 2014, Revised 3 November 2014, Accepted 5 January 2015, Available online 15 January 2015.
论文官网地址：https://doi.org/10.1016/j.patcog.2015.01.004
