Abstract-In order to solve the problem of lack of targeted guidance for learners in the community learning process, an improved algorithm for constructing communities was proposed. The algorithm used P2P architecture. Through the improved learning rules, the links between learners with similar interests were continuously strengthened. The community structure was automatically adjusted. In the end, learners with similar interests were grouped together. Taking university computer basic courses as an example, a self-organizing community learning system based on p2p network-based university computer basic courses was constructed. The results showed that compared with the traditional P2P architecture learning algorithm, this algorithm achieved better community construction quality and higher community construction speed. Therefore, this method can improve the learning effect of the university computer basic curriculum community.