The explosion of digital data in volume and complexity available through internet has driven a revolution in the way to handle large flows of information. This challenge has developed into new opportunities in many domains like health, transport, security, tourism, or e-commerce, with a blooming of new applications. Computers changed the way to do business few decades ago and the new challenge is to exploit large and diverse amounts of data. In fact, we refer to Big Data when the volume itself becomes part of the problem, when available techniques are not efficient enough. Scientific research in general, and Astronomy in particular, should of course benefit of the new data handling tools, like it did with computers before, and, e.g., deep learning, an active research area in machine learning and paIern recognition, offers excellent opportunities.

Astronomy is indeed a paradigm case for Big Data science. The continuing development of ground and space-based observatories, including large sky surveys, brings Astronomy to the Big Data era. Gaia or Euclid are examples in space but new groundbased projects, like LSST or SKA, will need the new tools even more. Means and methodologies to do research with these facilities will have, no doubt, to be evaluated.

Young astronomers will be more and more involved in the use and development of Big Data science for their research, not only mining data bases to get answers, but also producing complex simulations or even finding out new questions by recognising unexpected paIerns in the data, in other words, moving from a model-driven to adata-driven approach. Moreover, the traditional way of asking for observing time to investigate specific targets will also be drastically modified.

The purpose of this ISSI Forum is to convene a number of experts in the use of Big Data science for Astronomy with the aim of reflecting on the benefits and challenges of this new research tool.

February 5, 2019