http://gaia.fdi.ucm.es/research/lorecommender
LORecommender is a framework for the rapid prototyping of knowledge-based recommender systems applied to learning object recommendation. With a recommendation scheme of five stages as starting point, the framework can be configured and adapted to build different recommenders. The framework not only provides default implementations of alternative strategies for each stage, but can easily be extended with new implementations.