Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.adass.org/adass/proceedings/adass94/gulatir.ps Äàòà èçìåíåíèÿ: Tue Jun 13 20:47:22 1995 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 01:19:36 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: star

Astronomical Data Analysis Software and Systems IV
ASP Conference Series, Vol. 77, 1995
R. A. Shaw, H. E. Payne, and J. J. E. Hayes, eds.
Automated Classification of a Large Database of Stellar
Spectra
R. K. Gulati and R. Gupta
IUCAA, Post Bag 4, Ganeshkhind, Pune 411 007, India
P. Gothoskar and S. Khobragade
NCRA, TIFR Center, P.O. Box 3, Pune 411 007, India
Abstract. An Artificial Neural Network (ANN) is a versatile tool which
has been used both in academic research and industrial applications. In
astronomy, this technique has been used for a variety of applications, such
as telescope adaptive optics, classifying galaxies, and separating stars
from galaxies. The classification of a large database of stellar spectra,
which would be a Herculean task for human classifiers if done visually, is
an ideal problem for the ANN technique, which can handle such problems
without manual intervention. Recently, increased computational power,
combined with improvement in the ANN techniques, has provided an
efficient way to perform automatic classification.
We have implemented ANN to classify stellar spectra from large spec­
tral databases. We present here the Multilayer Back Propagation Net­
work (MBPN), which is used to classify stellar spectra obtained in the
optical and ultraviolet regions. The performance of MBPN shows that
the ANN is capable of classifying ultraviolet stellar spectra to an accu­
racy of about one spectral subclass for most of the cases. The scope of
this technique is expected to be expanded with the availability of large
homogeneous digitized stellar spectral databases.
1. Introduction
Progress in ground based and space instrumentation has brought us to a new
era of spectroscopy, where a large quantity of good quality stellar spectra has
started becoming available through well organized data centers. In order to
analyze these spectra and extract useful physical information about stars and
stellar systems, we need to develop fast and accurate methods. One way to
analyze these spectra is to classify them in terms of common visible properties.
Spectral classification, which conventionally has been done by human clas­
sifiers (Houk 1983; Houk & Smith­Moore 1988), involves large, time­consuming
efforts. We now require automation of the classification process. The main
advantages of automated over human classification is not only the speed with
which it can be done, but also accuracy, detection of variability, the elimination
of personal error, and the possibility of classification of higher dimensionality.
1

2
We (Gulati et al. 1994) have initiated a project to implement automated
classification schemes to digitized databases of optical and UV spectra by using
conventional metric distance minimization methods and Artificial Neural Net­
works. We have been using the Multilayer Back Propagation Network (MBPN).
Similar efforts have also been employed by another group to classify the stellar
spectra of high­dispersion objective prism plates using a neural network scheme
(von Hippel et al. 1994). ANN has also been applied to classify a near infra­red
database (Weaver 1994).
Here we present the performance of the ANN scheme on libraries of optical
and ultraviolet stellar spectra by comparing classifications determined by ANN
with those of human classifiers (i.e., catalog classifications).
2. Input and Pre­processing of Optical & UV Data
The optical data were taken from Silva (Silva & Cornell 1992) and Jacoby (Ja­
coby et al. 1984) libraries. A set of 55 spectra selected from the former library
was treated as the template database, and the test database was a set of 158
spectra from the latter library. Both sets were brought to a uniform wavelength
range of 3510--6800 š A with 5 š A sampling and 11 š A resolution, and normalized
to a value of 100 at 5450 š A . Instead of using the full spectral information, a
set of 161 wavelength positions was used to monitor the fluxes which are diag­
nostic of the spectral classes as given by human experts (Jaschek and Jaschek
1990). Catalog classifications of the spectra were taken from the respective
libraries. The spectra covered stars of solar metallicity, types O--M, and lu­
minosity classes I--V. Each spectro­luminosity class was coded with a number
x = (1000 \Theta A1+100 \Theta A2+(1:5+2 \Theta A3)) , where A1 was the main spectral type
of the star (i.e., O to M types coded from 0.0 to 9.5), A2 was the sub­spectral
type (coded from 0.0 to 9.5) and A3 the luminosity class (i.e., classes I to V
coded as 0 to 4). For example, a B2I star and a G9.5V star would be coded as
2201.5 and 5959.5, respectively.
The input database for the UV data was the IUE Low Resolution Spectra
(Heck et al. 1984). A set of 128 spectra spanning 75 spectro­luminosity classes
was selected as the template and another set of 83 spectra was used as the
test set. The catalog classification was taken from this catalog, where like MK
classification, the UV classification is given as O, B, F, etc., as main classes,
subclasses ranging from 0.0 to 9.5, and luminosity classes represented as s, g,
and d for super­giants, giants, and dwarfs, respectively. The wavelength range
of the UV spectra is 1213--3201 š A with 2 š A sampling and 6 š A resolution. The
spectra were monitored at 35 wavelength positions which are the diagnostic of
these spectral classes as given in Table 1 of the IUE catalog (Heck et al. 1984).
Here, too, the spectral coding was done by using a number x = (1000 \Theta A1 +
100 \Theta A2 + A3) , where A1 was the main spectral type of the star (i.e., O to
F types coded as 1 to 4), A2 was the sub­spectral type of the star (coded from
0.0 to 9.5), and A3 the luminosity class of the star (i.e., 2, 5, or 8 for s, g, or
d). For example, stars dB2.5, gO9.5 and sF7 (ultraviolet classes) were coded as
2258, 1955, and 4702, respectively.

3
3. The ANN Architectures
We used the standard feedforward supervised neural network, known as ``multi­
layer backpropagation network (MBPN)'' (Rumelhart et al. 1986), for classifying
the databases of optical and ultraviolet stellar spectra into different classes of
stars. As mentioned earlier, the number of output classes was 55 in the op­
tical case and 75 in the UV case. The input data points for optical and UV
data classifiers were different, so the ANN classifiers were selected with different
architectures for optical and UV data. The optical classifier was found to be
optimal with the configuration 161:64:64:55 and the UV classifier was config­
ured as 35:71:75. The configuration numbers show the input size, hidden nodes,
and, at the end, the output nodes, respectively. Once the training was over, the
networks could classify the large databases of stellar spectra within a minute,
without any human intervention.
4. Performance
The performance of the ANN technique can be judged from Figure 1, which
shows the 3D plots of classification errors in luminosity and spectral type on x
and y axes and the percentage of total test sample along the z axis, respectively,
for optical and UV data. In these plots an ideal classification would appear as
a single peak of 100% value in the center of the (x; y) plane, signifying that all
spectra are classified correctly with no errors in either luminosity or spectral
type. One sub­spectral type error means 100 units error along the y­axis of
these 3D plots. Statistical parameters, such as linear correlation coefficients and
standard deviations, were computed on the scatter plots (for details see Gulati
et al. 1994), and it was found that the classification error for optical was about
two subclasses and for UV it is about one subclass, barring a few stars which
clearly show more than 100 units of error. These stars require further detailed
studies.
5. Conclusions and Future Steps
We do not see any gross mis­classification with the automated ANN scheme
and the schemes are quite efficient for large databases. However, we feel that
a more homogeneous and complete database is required for the ANN training
to perform better. The implementation of ANN on a parallel computer would
significantly reduce the training time. In the future, we plan to use a library
of synthetic spectra based on stellar atmosphere models (Gulati et al., 1993) to
tag information on the stellar physical parameters.
Acknowledgments. R. K. Gulati wishes to acknowledge the generous fi­
nancial support from the organizing committees of ADASS IV, which allowed
him to present this paper at the conference.
References
Gulati, R. K., Malagini, M. L., & Morossi, C. 1993, ApJ, 413, 166

4
O P T C A L
­5
0
5
X
­200
­100
0
100 200
Y
0
5
10
%
­5
0
5
X
­200
­100
0
100 200
Y
U V
­5
­2.5
0
2.5
5
X
­200
­100
0
100 200
Y
0
10
20
30
%
­5
­2.5
0
2.5
5
X
­200
­100
0
100 200
Y
Figure 1. 3D plots for classification errors in luminosity (x­axis) and
spectral type (y­axis) vs. the % of total number of test spectra for
Optical and UV data.
Gulati, R. K., Gupta, R., Gothoskar, P., & Khobragade, S. 1994, ApJ, 426, 340
Heck, A., Egret, D., Jaschek, M., & Jaschek, C. 1984, in IUE Low­Resolution
Spectra: A Reference Atlas--Part I, Normal Stars, ESA SP­1052
Houk, N. 1983, in The MK Process and Stellar Classification, ed. R. F. Garrison,
(Toronto, David Dunlop Observatory), p. 85
Houk, N., & Smith­Moore, M. 1988, in University of Michigan Catalogue of
Two­Dimensional Spectral Types for the HD Stars, 1988, Vol. 4
Jacoby, G. H., Hunter, D. A., & Christian, C. A. 1984, ApJS, 56, 257
Jaschek, C., & Jaschek, M. 1990, The Classification of Stars, (Cambridge, Cam­
bridge Univ. Press)
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. 1986, Nature, 323, 533
Silva, D. R., & Cornell, M. E. 1992, ApJS, 81, 865
von Hippel, T., Storrie­Lombardi, L. J., Storrie­Lombardi, M. C. & Irwin, M. J.
1994, MNRAS, 269, 97
Weaver, Wm. B. 1994, in The MK Process at 50 Years: A Powerful tool for
Astrophysical Insight, ASP Conf. Series, Vol. 60, eds. C. J. Corbally,
R. O. Gray, and R. F. Garrison (San Francisco, ASP), p. 303