Synapses of active galactic nuclei:. Comparing X-ray and optical classifications using artificial neural networks

Context. Many classes of active galactic nuclei (AGN) have been defined entirely through optical wavelengths, while the X-ray spectra have been very useful to investigate their inner regions. However, optical and X-ray results show many discrepancies that have not been fully understood yet.
Aims: The main purpose of the present paper is to study the synapses (i.e., connections) between X-ray and optical AGN classifications.
Methods: For the first time, the newly implemented efluxer task allowed us to analyse broad band X-ray spectra of a sample of emission-line nuclei without any prior spectral fitting. Our sample comprises 162 spectra observed with XMM-Newton/pn of 90 local emission line nuclei in the Palomar sample. It includes, from the optical point of view, starbursts (SB), transition objects (T2), low-ionisation nuclear emission line regions (L1.8 and L2), and Seyfert nuclei (S1, S1.8, and S2). We used artificial neural networks (ANNs) to study the connection between X-ray spectra and optical classes.
Results: Among the training classes, the ANNs are 90% efficient at classifying the S1, S1.8, and SB classes. The S1 and S1.8 classes show a negligible SB-like component contribution with a wide range of contributions from S1- and S1.8-like components. We suggest that this broad range of values is related to the high degree of obscuration in the X-ray regime. When including all the objects in our sample, the S1, S1.8, S2, L1.8, L2/T2/SB-AGN (SB with indications of AGN activity in the literature), and SB classes have similar average X-ray spectra, but these average spectra can be distinguished from class to class. The S2 (L1.8) class is linked to the S1.8 (S1) class with a larger SB-like component than the S1.8 (S1) class. The L2, T2, and SB-AGN classes constitute a class in the X-rays similar to the S2 class, albeit with larger portions of SB-like component. We argue that this SB-like component might come from the contribution of the host galaxy emission to the X-rays, which is high when the AGN is weak. Up to 80% of the emission line nuclei and, on average, all the optical classes included in our sample show a significant fraction of S1-like or S1.8-like components. Thus, an AGN-like component seems to be present in the vast majority of the emission line nuclei in our sample.
Conclusions: The ANN trained in this paper is not only useful for studying the synergies between the optical and X-ray classifications, but might also be used to infer optical properties from X-ray spectra in surveys like eRosita.

Table 1 and Appendices are available in electronic form at http://www.aanda.org