Optimal randomized classification trees

Rafael Blanquero,Emilio Carrizosa,Cristina Molero-Río, Dolores Romero Morales

Computers & Operations Research（2021）

引用 0|浏览0

暂无评分

摘要

•We introduce ORCT, a continuous optimization approach to build classification trees.•Accuracy is competitive against CART and recent integer programming-based approaches.•ORCT naturally provides class membership probabilities tailored to each individual.•Preferences on classification accuracy in some classes are easily handled.

查看译文

关键词

Classification and regression trees,Cost-sensitive classification,Nonlinear programming

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要