Table A.1
Percentage of False Positives and False Negatives for the different strategies of train-test splits and threshold values.
| Equal Ratio (%) | Balance with 0.5 (%) | Balance with 0.8 (%) | Unbalance (%) | |
|---|---|---|---|---|
| False Positive | 2 | 14 | 9 | 3 |
| False Negative | 17 | 5 | 9 | 18 |
Notes. False positive (galaxy) and false negative (QSO2) comparison for the XGBoost algorithm: Equal ratio training set, balanced training set with threshold 50% and 80%, and unbalanced training set. The results are presented in percentage based on the total number of objects in each classification class.
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.