This paper proposes a personalized explicit- and tacit-knowledge recommendation model for a virtual research community. The proposed model aims to recommend both useful journal documents (explicit knowledge) and community members who can discuss the information (tacit knowledge) on-line in real time. A middleware system, the personalized knowledge recommender (PKR) system, that was constructed from this model is presented. The model combines content-based (CB) and collaborative filtering (CF) methods, to make explicit and tacit knowledge recommendations. Unlike other similar systems, this system adapts CB and CF in different ways to provide users with not only “interest-related” documents for reference but also connections to the “related knowledge owners” for further on-line discussion. An e-journal paper recommendation for a virtual research community is used as an example to evaluate the performance of PKR in terms of mean absolute error, precision, recall and F-measure. PKR creates the specialty of explicit- and tacit-knowledge recommendations.