| Popis: |
This thesis describes the cross-matching software solution lying at thecore of the computational infrastructure for the VASCO project. The VASCO project has a goal to mine historical all-sky surveys to find astronomical anomalies. It aims to give new clues to either SETI research or in theoretical astrophysics, and serves as a starting point for observational followup and/or new theoretical developments. Cross-matching throughout the thesis refers to comparing billions of astronomical objects recorded in the historical USNO and PanSTARRS allskysurveys. This thesis describes how to approach this huge computational challenge using methods of big data and cloud computing. The techniques described in this thesis resulted in a list of about 400 thousand objects which are usable in further analysis in the machine learning tool called ML-Blink. |
