Identifying type II quasars at intermediate redshift with few-shot learning photometric classification

Open Access

Table 2

Classification evaluation metrics for KNNeigbors, RandomForest, LightGBM, CatBoost, XGBoost, MLP, and generalised stacking with a threshold ≥0.5.

Algorithm	Accuracy	Precision	Recall	F1-Score	F1-Score, cv=10
`KNNeigbors`	0.879	0.897	0.887	0.879	0.876 ± 0.063
`RandomForest`	0.879	0.881	0.882	0.879	0.867 ± 0.080
`XGBoost`	0.966	0.965	0.965	0.965	0.943 ± 0.041
`CatBoost`	0.966	0.966	0.968	0.965	0.926 ± 0.041
`LightGBM`	0.948	0.949	0.949	0.948	0.921 ± 0.046
`MLP`	0.948	0.950	0.952	0.948	0.921 ± 0.035
Generalised stacking	0.966	0.966	0.968	0.965	–

Notes. The final column shows the F1-Score metric using k-fold cross-validation, with k = 10, that allows the standard deviation for the final classifications.

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.