We present in this paper a framework to model information diffusion in social networks based on linear multivariate Hawkes processes and the latent Dirichlet allocation topic model. Our model exploits the effective broadcasting times of information by users and a fuzzy scheme of information dissemination, where users broadcast information as a mixture of different latent topics, guaranteeing thus a more realistic view of the information diffusion process. The proposed model takes into consideration not only interactions between users but also interactions between topics, which provides a deeper analysis of influences in social networks. We provide an estimation algorithm based on nonnegative matrix factorization techniques, which are coupled together with a modified collapsed Gibbs sampler and a modified variational Bayes method. This allows a more data-driven estimation of hidden influences in social networks