Predictive model of prognosis index for invasive micropapillary carcinoma of the breast based on machine learning: a SEER population-based study

Table 3 Model performance for the 5-year survival

Algorithms	Accuracy	Speificity	Precision	PPV	NPV	AUC^a
SVM-train	0.868	0.992	0.912	0.912	0.864	0.936
SVM-test	0.863	0.989	0.889	0.889	0.861	0.814
KNN-train	0.859	0.971	0.761	0.761	0.870	0.861
KNN-test	0.829	0.968	0.667	0.667	0.843	0.743
RandomForest-train	0.959	0.989	0.949	0.949	0.962	0,986
RandomForest-test	0.863	0.947	0.706	0.706	0.890	0.787
ExtraTrees-train	0.964	1.000	1.000	1.000	0.957	0.994
ExtraTrees-test	0.846	0.936	0.647	0.647	0.880	0.751
XGBoost-train	0.878	0.984	0.867	0.867	0.880	0.931
XGBoost-test	0.863	0.979	0.818	0.818	0.868	0.863

ISSN: 1472-6947