Free Access

Table 2.

Accuracy (in %) of the supervised machine learning methods for the automated binary morphological classification (total, for early E and late L morphological types, rms error) of modified training sample of 11301 galaxies from the SDSS DR9 at z < 0.1 (region with the overlap of types in Figs. 1 and 3).

Classifier vs. accuracy Total E type L type Error
Naive Bayes 66.8 64.1 70.4 ±1.2
K-nearest neighbors 79.4 80.3 78.6 ±0.7
Logistic regression 81.9 83.9 80.3 ±0.3
Random forest 82.4 87.6 78.6 ±0.4
Support-vector machine 84.3 89.0 80.6 ±0.5

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.