Recommending Software Architecture for Customized Learning Management System

In this work we have studied the learning management systems (LMSs) and proposed a reference model, serving to both analysis and synthesis of LMSs. Such model has been implemented and evaluated, being composed of two representations and facilities specially oriented to two important actors in LMS development. These actors are the author and the developer, where a concept map is offered to the first one in order to help authors to select services that will compose the intended LMS. The flexible reference architecture is dedicated to the software developers in order to help them to implement the user's requirements. In this paper, we propose a knowledge-based recommender system to automatically approximate these two representations through rules that map the needs of the author in design decisions to developers. Hence, the system may recommend a new architecture for customized LMS or even an existing LMS to the author. The proposed system has been evaluated in the task of constructing a specific LMS thus showing positive results.