Cross-checking the reliability of various stellar spectral databases is an important and desirable exercise. Since number of stars in various databases have no known spectral types and some of the libraries do not have complete coverage resulting in gaps. We use an automated classification scheme based on Artificial Neural Networks (ANN) to cross-classify stars in the Indo-US stellar spectral library (Valdes et al. 2004), JHC (Jacoby, Hunter & Christian 1984), ELODIE spectra (Moultaka et al. 2004) and STELIB (Le Borgne et al. 2003). We have also examined the effects of over-training and over-fitting on the classification efficiency of a Neural Network. It is hoped that such a automated data analysis and validation technique will be useful in the future.