.png){kind=icon}
返回
Mathematical programming model mining: A systematic field survey
.png){kind=icon}
PRE
AI
DOI：10.1016/j.cosrev.2026.100905.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
• The technology for mining Mathematical Programming (MP) models from data rarely exceeds TRL 3. • Most discovery approaches rely on exemplary solutions and do not scale to hundreds of examples. • Most conformance checking and enhancement algorithms are based on features of the given MP model, ignoring domain knowledge. • High-level modeling languages, structuring variables and examples into tensors, and limits on constraint size are promising research directions. • The field lacks standard conformance measures of fitness, precision, and generalization of MP models w.r.t. domain knowledge. • The field requires standardized benchmarks and datasets for MP model mining problems. • The field needs parameterized simulators of human domain experts.
期刊
IF:
12.7
论文数:
2.3K
被引数:
5.2K
机构
暂无机构信息
引用论文
.png){kind=icon}
分享收藏分享收藏
One-class synthesis of constraints for Mixed-Integer Linear Programming with C4.5 decision trees
PREAI
分享收藏
查看更多内容