The virtual community established by online forums is the most typical, mature, and representative concrete form in the online society. The knowledge recommendation based online forum system is a type of thematic forum that provides users with a richer and more diverse knowledge experience by launching high-quality content that meets their needs. Firstly, the characteristics of openness, anonymity, and interactivity of online forums were analyzed; Next, this article constructs a knowledge recommendation process consisting of user preference model construction, knowledge model construction, and recommendation algorithm design; Then, a collaborative filtering algorithm process was designed, and a mathematical model was constructed based on the process of constructing a user rating matrix, searching for similar users, and generating recommendation results; Finally, this article designs an overall system architecture consisting of user operation layer, system function layer, business logic layer, and data persistence layer. The research results provide a complete solution for the development of knowledge recommendation based online forum systems.