Data Mining and Discovery of Astronomical Knowledge

• Published in: Scientific Data Mining and Knowledge Discovery pp. 319 - 341
• Main Author: Al-Naymat, Ghazi
• Format: Book Chapter
• Language: English
• Published: Berlin, Heidelberg Springer Berlin Heidelberg 2010
• Subjects: Main Galaxy; Maximal Clique; Neighborhood List; Object Type; Spatial Object
• ISBN: 3642027873; 9783642027871
• DOI: 10.1007/978-3-642-02788-8_12

Spatial data is essentially different from transactional data in its nature. The objects in a spatial database are distinguished by a spatial (location) and several non-spatial (aspatial) attributes. For example, an astronomy database that contains galaxy data may contain the x, y and z coordinates (spatial features) of each galaxy, their types and other attributes. Spatial datasets often describe geo-spatial or astro-spatial (astronomy related) data. In this work, we use a large astronomical dataset containing the location of different types of galaxies. Datasets of this nature provide opportunities and challenges for the use of data mining techniques to generate interesting patterns. One such pattern is the co-location pattern. A co-location pattern is a group of objects (such as galaxies) each of which is located in the neighborhood (within a given distance) of another object in the group.