Identification of metal-poor stars among field stars is extremely useful for studying the structure and evolution of the Galaxy and of external galaxies. We search for metal-poor stars using the artificial neural network (ANN) and extend its usage to determine absolute magnitudes. We have constructed a library of 167 medium-resolution stellar spectra (R~1200) covering the stellar temperature range of 4200 to 8000K, logg range of 0.5 to 5.0, and [Fe/H] range of -3.0 to dex. This empirical spectral library was used to train ANNs, yielding an accuracy of 0.3dex in [Fe/H], 200K in temperature, and 0.3dex in logg. We found that the independent calibrations of near-solar metallicity stars and metal-poor stars decreases the errors in Teff and logg by nearly a factor of two.

Cone search capability for table J/A+A/556/A121/table1 (List of observed stars and their parameters)

Cone search capability for table J/A+A/556/A121/table2 (Estimated atmospheric parameters for candidate metal-poor stars)