Improving the discovery of near-Earth objects with machine-learning methods

Open Access

Table 5

Confusion matrices for filters 0–2 using 2019–2023 NEOCP data.

True label	Predicted Label
	Filter 0			Filter 1			Filter 2
	Non-NEO	Unclassified	Total	Non-NEO	Unclassified	Total	Non-NEO	Unclassified	Total
NEO	0	15591	15591	9	15582	15591	14	15577	15591
Non-NEO	911	7085	7996	1362	6634	7996	1557	6439	7996

Notes. Most of the data are not classified, but the precision of classifying non-NEOs among the classified sample is very high (≫99%).

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.