* <strong>过拟合</strong>：就是在训练集上准确率非常高，而在测试集上准确率低。 * <strong>分类问题评价方法</strong>：[https://www.zhihu.com/question/30643044](https://www.zhihu.com/question/30643044) * <strong>分类模型评估之ROC-AUC曲线和RPC曲线</strong>：[https://blog.csdn.net/pipisorry/article/details/51788927](https://blog.csdn.net/pipisorry/article/details/51788927) * <strong>混淆矩阵记忆口诀</strong>：包含主对角线（真正、真反），剩余（假正，假反）。  * [模型评估方法（混淆矩阵、查准率&查全率&P-R图）](https://blog.csdn.net/zzh1301051836/article/details/88965040) * [欠拟合的解决方案有哪些？](https://support.huaweicloud.com/modelarts_faq/modelarts_05_0170.html) * scikit-learn算法选择路径：  * [推荐系统中SVD算法详解](https://blog.csdn.net/fool_ran/article/details/79384040) * [scikit-learn算法选择路径解释](https://blog.csdn.net/hjwbit/article/details/88065566?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromBaidu-1.control&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromBaidu-1.control) * [python 机器学习：sklearn全景图](https://blog.csdn.net/huoyingchong64/article/details/89879134?utm_medium=distribute.pc_relevant_t0.none-task-blog-BlogCommendFromMachineLearnPai2-1.control&depth_1-utm_source=distribute.pc_relevant_t0.none-task-blog-BlogCommendFromMachineLearnPai2-1.control) * [scikit-learn文档](https://sklearn.apachecn.org/docs/master/11.html)