DECISION TREE CLASSIFIERS FOR STAR/GALAXY SEPARATION

Journal Article · Wed Jun 15 00:00:00 EDT 2011 · Astronomical Journal (New York, N.Y. Online)

DOI:https://doi.org/10.1088/0004-6256/141/6/189· OSTI ID:21583105

Vasconcellos, E C; Ruiz, R S. R. ^[1]; De Carvalho, R R; Capelato, H V ^[2]; Gal, R R ^[3]; LaBarbera, F L ^[4]; Frago Campos Velho, H ^[5]; Trevisan, M ^[6]

---

1. CAP, National Institute of Space Research, Av. dos Astronautas 1758, Sao Jose dos Campos 12227-010 (Brazil)
2. DAS, National Institute of Space Research, Av. dos Astronautas 1758, Sao Jose dos Campos 12227-010 (Brazil)
3. Institute for Astronomy, University of Hawaii, 2680 Woodlawn Dr., Honolulu, HI 96822 (United States)
4. INAF-Osservatorio Astronomico di Capodimonte, via Moiariello 16, Napoli 80131 (Italy)
5. LAC, National Institute of Space Research, Av. dos Astronautas 1758, Sao Jose dos Campos 12227-010 (Brazil)
6. IAG, University of Sao Paulo, Rua do Matao 1226, Sao Paulo 05508-090 (Brazil)

We study the star/galaxy classification efficiency of 13 different decision tree algorithms applied to photometric objects in the Sloan Digital Sky Survey Data Release Seven (SDSS-DR7). Each algorithm is defined by a set of parameters which, when varied, produce different final classification trees. We extensively explore the parameter space of each algorithm, using the set of 884,126 SDSS objects with spectroscopic data as the training set. The efficiency of star-galaxy separation is measured using the completeness function. We find that the Functional Tree algorithm (FT) yields the best results as measured by the mean completeness in two magnitude intervals: 14 {<=} r {<=} 21 (85.2%) and r {>=} 19 (82.1%). We compare the performance of the tree generated with the optimal FT configuration to the classifications provided by the SDSS parametric classifier, 2DPHOT, and Ball et al. We find that our FT classifier is comparable to or better in completeness over the full magnitude range 15 {<=} r {<=} 21, with much lower contamination than all but the Ball et al. classifier. At the faintest magnitudes (r > 19), our classifier is the only one that maintains high completeness (>80%) while simultaneously achieving low contamination ({approx}2.5%). We also examine the SDSS parametric classifier (psfMag - modelMag) to see if the dividing line between stars and galaxies can be adjusted to improve the classifier. We find that currently stars in close pairs are often misclassified as galaxies, and suggest a new cut to improve the classifier. Finally, we apply our FT classifier to separate stars from galaxies in the full set of 69,545,326 SDSS photometric objects in the magnitude range 14 {<=} r {<=} 21.

• OSTI ID: 21583105
• Journal Information: Astronomical Journal (New York, N.Y. Online), Vol. 141, Issue 6; Other Information: DOI: 10.1088/0004-6256/141/6/189; ISSN 1538-3881
• Country of Publication: United States
• Language: English

Related Subjects

79 ASTROPHYSICS
COSMOLOGY AND ASTRONOMY
ALGORITHMS
CATALOGS
CLASSIFICATION
DATA ANALYSIS
DECISION TREE ANALYSIS
GALAXIES
STARS
DOCUMENT TYPES
MATHEMATICAL LOGIC