J/MNRAS/353/211 Method for selection of quasars (Carballo+, 2004)
Selection of quasar candidates from combined radio and optical surveys using
neural networks.
Carballo R., Cofino A.S., Gonzalez-serrano J.I.
<Mon. Not. R. Astron. Soc., 353, 211-220 (2004)>
=2004MNRAS.353..211C 2004MNRAS.353..211C
ADC_Keywords: QSOs
Keywords: methods: data analysis - methods: statistical - quasars: general
Abstract:
The application of supervised artificial neural networks (ANNs) for
quasar selection from combined radio and optical surveys with
photometric and morphological data is investigated, using the list of
candidates and their classification from the work of White et al.
(2000, Cat. J/ApJS/126/133>) Seven input parameters and one output,
evaluated to 1 for quasars and 0 for non-quasars during the training,
were used, with architectures 7: 1 and 7: 2: 1. Both models were
trained on samples of 800 sources and yielded similar performance on
independent test samples, with reliability as large as 87 per cent at
80 per cent completeness (or 90 to 80 per cent for completeness from
70 to 90 per cent). For comparison, the quasar fraction from the
original candidate list was 56 per cent.
Description:
Predictions of the probabilities for the 98 candidates without
spectroscopic classification in White et al. (2000, Cat.
J/ApJS/126/133>) are presented and compared with the results from
their work. The values obtained for the two ANN models and the
decision trees are found to be in good agreement.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table4.dat 45 98 Quasar probabilities for the 98 candidates
without spectroscopic classification in
White et al. (2000, Cat. J/ApJS/126/133)
--------------------------------------------------------------------------------
See also:
J/ApJS/126/133 : The FIRST bright quasar survey. II. (White+, 2000)
Byte-by-byte Description of file: table4.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 15 A15 --- FIRSTJ FIRST J designation (based on J2000.0 position)
17- 20 F4.2 --- P7-1 Quasar probability from ANN 7:1 method (1)
23- 26 F4.2 --- e_P7-1 rms uncertainty on P7-1
29- 32 F4.2 --- P7-2-1 Quasar probability from ANN 7:2:1 method (1)
35- 38 F4.2 --- e_P7-2-1 rms uncertainty on P7-2-1
41- 44 F4.2 --- POC1 Quasar probability from oblique decision tree
classifier OC1 (Murthy et al., 1994,
J. Artif. Intell. Res., 2, 1)
--------------------------------------------------------------------------------
Note (1): We used two different ANN architectures.
The first one, denoted as 7:1, does not include hidden layers
and it is also known as a logistic discrimination model.
The second architecture includes a hidden layer with two nodes,
and it is denoted as 7:2:1.
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Patricia Vannier [CDS] 14-Mar-2005