In the context of the provision of library services, bibliomining is defined as the process of applying
data mining techniques to extract patterns of behavior from library databases. The application of bibliomining
has hitherto focused on the optimal allocation of budgetary resources based on the past circulation of
documents. There is no specific application of bibliomining methods to the recognition of user behavior
patterns in the available literature. This paper aims to bridge this gap and presents the application of clustering
methods on real data from a public library to obtain behavioral patterns of representative users. The results of
this work can then be implemented to achieve more efficient management of public libraries. This paper
develops a method of bibliomining, including the characteristics of the various stages of the process.
Furthermore, the whole process is applied to research conducted in 2012 in the largest public library in the
Czech Republic – the Municipal Library of Prague. The results are interpreted and a proposal for continuation
of research is also included.