Nowadays, the information overload highlights the importance of personalization systems providing services according to users' interests and needs, as expressed by their User Models (UMs). Since the quality of the provided personalized depends on the accuracy of the UMs, the services would benefit from enriching their UMs through mediating partial UMs, built by other services. This paper elaborates on cross-technique mediation of the UMs from collaborative to content-based services using an external Knowledge Base. Experimental evaluation conducted in the domain of movies, shows that for small UMs, the personalization provided using the resulting content-based UMs, outperforms the personalization provided using the original collaborative UMs.