Telco Customer Churn基准的信任性审计：99%准确率隐藏成本

1. 12:00
   arXiv · Machine Learning Theory精选

   99% 准确率的隐藏成本：电信客户流失基准的可信度审计

   论文审计了 IBM Telco 客户流失基准（n=7,043），发现预拆分 SMOTE 使流失类 F1 虚高 13.1 个百分点，约 36% 合成训练点是测试集实例的最近邻插值；TotalCharges 与 tenure×MonthlyCharges 的 R²=0.999，移除后准确率变化不足 0.2 个百分点但 TreeSHAP 将其排在第九位，误导解释。

Current heat 9·Comparable peak 10（Oct 2）·Comparable change over 24 hours –

The trend compares only the same participants observed continuously; its range may be smaller than the current heat count. Move or click on the chart to inspect hourly heat; use the left and right arrow keys to switch.