FleBiC: Learning classifiers from high-dimensional biomedical data using discriminative biclusters with non-constant patterns
作者：
Highlights：
• Discriminative non-constant patterns (additive, multiplicative, order-preserving) aid high-dimensional data classification.
• In biomedicine, non-constant assumptions handle variable individual biophysiology, disease morphology and progression stage.
• Biclustering large sets of noise-tolerant discriminative patterns minimizes match scarcity, emulating boosting principles.
• Coherence-sensitive pattern scoring (training) and matching (testing) improve associative classification.
• FleBiC offers a way out of generalization difficulties, places statistical guarantees, and promotes interpretability.
摘要
•Discriminative non-constant patterns (additive, multiplicative, order-preserving) aid high-dimensional data classification.•In biomedicine, non-constant assumptions handle variable individual biophysiology, disease morphology and progression stage.•Biclustering large sets of noise-tolerant discriminative patterns minimizes match scarcity, emulating boosting principles.•Coherence-sensitive pattern scoring (training) and matching (testing) improve associative classification.•FleBiC offers a way out of generalization difficulties, places statistical guarantees, and promotes interpretability.
论文关键词：Associative classification,Discriminative paterns,Biclustering,Non-constant patterns,Biomedical data,High-dimensional data
论文评审过程：Received 9 February 2020, Revised 30 January 2021, Accepted 6 February 2021, Available online 20 February 2021, Version of Record 24 March 2021.
论文官网地址：https://doi.org/10.1016/j.patcog.2021.107900
