Classification and characterization of variable phenomena and transient phenomena are critical for astrophysics and cosmology. These objects are commonly studied using photometric time series or spectroscopic data. Given that many ongoing and future surveys are in time-domain and given that adding spectra provide further insights but requires more observational resources, it would be valuable to know which objects should we prioritize to have spectrum in addition to time series. We propose a methodology in a probabilistic setting that determines a-priory which objects are worth taking spectrum to obtain better insights, where we focus 'insight' as the type of the object (classification). Objects for which we query its spectrum are reclassified using their full spectrum information. We first train two classifiers, one that uses photometric data and another that uses photometric and spectroscopic data together. Then for each photometric object we estimate the probability of each possible spectrum outcome. We combine these models in various probabilistic frameworks (strategies) which are used to guide the selection of follow up observations. The best strategy depends on the intended use, whether it is getting more confidence or accuracy. For a given number of candidate objects (127, equal to 5% of the dataset) for taking spectra, we improve 37% class prediction accuracy as opposed to 20% of a non-naive (non-random) best base-line strategy. Our approach provides a general framework for follow-up strategies and can be extended beyond classification and to include other forms of follow-ups beyond spectroscopy.

We present the study of nineteen low X-ray luminosity galaxy clusters (L$_X \sim$ 0.5--45 $\times$ $10^{43}$ erg s$^{-1}$), selected from the ROSAT Position Sensitive Proportional Counters (PSPC) Pointed Observations (Vikhlinin et al. 1998) and the revised version of Mullis et al. (2003) in the redshift range of 0.16 to 0.7. This is the introductory paper of a series presenting the sample selection, photometric and spectroscopic observations and data reduction. Photometric data in different passbands were taken for eight galaxy clusters at Las Campanas Observatory; three clusters at Cerro Tololo Interamerican Observatory; and eight clusters at the Gemini Observatory. Spectroscopic data were collected for only four galaxy clusters using Gemini telescopes. With the photometry, the galaxies were defined based on the star-galaxy separation taking into account photometric parameters. For each galaxy cluster, the catalogues contain the PSF and aperture magnitudes of galaxies within the 90\% completeness limit. They are used together with structural parameters to study the galaxy morphology and to estimate photometric redshifts. With the spectroscopy, the derived galaxy velocity dispersion of our clusters ranged from 507 km~s$^{-1}$ for [VMF98]022 to 775 km~s$^{-1}$ for [VMF98]097 with signs of substructure. Cluster membership has been extensively discussed taking into account spectroscopic and photometric redshift estimates. In this sense, members are the galaxies within a projected radius of 0.75 Mpc from the X-ray mission peak and with cluster centric velocities smaller than the cluster velocity dispersion or 6000 km~s$^{-1}$, respectively. These results will be used in forthcoming papers to study, among the main topics, the red cluster sequence, blue cloud and green populations; the galaxy luminosity function and cluster dynamics.