Visual data mining of astronomic data with virtual reality spaces:: Understanding the underlying structure of large data sets

The information explosion in astronomy requires the development of data mining procedures that speed up the process of scientific discovery, and the in-depth understanding of the internal structure of the data. This is crucial for the identification of valid; novel, potentially useful, and understandable patterns (regularities; oddities, etc.). A Virtual Reality (VR) approach for large heterogeneous, incomplete and imprecise information is introduced for the problem of visualizing and analyzing astronomic data. The method is based on mappings between one heterogeneous space representing the data, kind a homogeneous virtual reality space. This VR-based visual data mining technique allows the incorporation of the unmatched geometric capabilities of the human brain into the knowledge discovery process, and helps in understanding dates structure and patterns. This approach has been applied successfully to a wide variety of real-world domains, and it has a large potential in astronomy. Examples are presented from the domain of galaxy research.