<p>Social trust is a valuable asset in modern recommender systems, enabling more personalized and accurate recommendations by aggregating the opinions of trusted users within social networks. Inferring implicit social trust from user interactions and behaviors is a complex task that requires significant computational resources. Large-scale social networks amplify this challenge. Moreover, collecting and analyzing personal data for this purpose raises privacy concerns. To improve computational efficiency and privacy, we propose a lightweight trust model that relies solely on users’ communication behavior, eliminating the need for in-depth content analysis. We consider that the propagation of information from one member to many other members indicates that a high degree of trust is placed in the information and, implicitly, in its source. By leveraging the constructed behavioral trust network, our cooperation-based source recommendation system outperforms traditional approaches that neglect trust relationships. Extensive experiments on real-world datasets validate the effectiveness of our approach in terms of accuracy and error minimization.</p>