摘要：We discuss the Personal Generative Library (PGL) concept that covers models to describe some structural, functional and managerial aspects. Since the concept, to some extent, was realized in our previous research, in this paper, we focus more on the managerial aspects. In this regard, we propose the feature model-driven approach to implement those aspects using meta-programming techniques. First, we present the basic idea and theoretical background of the approach. Then we discuss the PGL architecture, its functionality and management procedures that are supported by the developed meta-programs. We outline the process of designing meta-programs through the series transformations of feature models. The main contribution of the paper is the implementation of the concept itself that enables, to some extent, to resolve the well known problems: library scaling and excluding synonymy in search. Furthermore, we have extended the potential of generative reuse (meaning a higher extent of automation as compared to the component-based reuse) by applying it not only at the library entity level (a great deal of PGL items are generative learning objects (GLO)), but also at the whole library, i.e. its management level. Therefore, the approach enables the automatic formation of annotations for PGL entities and generation of queries to support managing procedures. We have approved the approach by presenting a case study and some experimental results.
