The domain model is an essential part of an adaptive learning system. For each educational course, it involves educational content and semantics, which is also viewed as a form of conceptual metadata about educational content. Due to the size of a domain model, manual domain model creation is a challenging and demanding task for teachers or content (and metadata) authors. We propose a method for the automated acquisition of hierarchical relationships between relevant domain terms from educational content, which constitutes a fundamental step in the semantic composition of an educational course. The method is based on existing text mining methods and applied to educational content. We evaluate our approach by performing several experiments. The evaluation shows that the method's performance is very promising. A study in a real-user scenario reveals that despite the fact that utilization of our method does not necessarily improve the speed of the domain model creation nor does it reduce the overall difficulty of the task, a significant improvement in the quality of resulting domain models has been observed. Our work is a promising contribution to the growing field of automated domain model acquisition.