Data mining techniques have long been used in astronomy for analyzing massive data sets obtained through observations. As a result, astronomy data sets have led to several interesting problems in the mining of scientific data. These problems are likely to become more challenging as the astronomy community brings several surveys online as part of the National Virtual Observatory, giving rise to the possibility of mining across surveys. In this article, we discuss our experiences with mining the FIRST survey to identify galaxies with a bent-double morphology. We describe the approach we took to mine the FIRST image and catalog data, the issues that we had to address in working with a real data set, and the lessons that we learned in the process of classifying bent-double galaxies.