Abstract: Objective Existing geography topic models do not consider the degree to which different regions influ-ence microblog topics.Meanwhile,these models describe the topic evolutions in a discrete mannerwhich prevents the acquisition of topic intensities over continuous time.This paper proposes a novelspatio-temporal topic model to discover microblog topics by introducing continuous time and region in-fluences.A city was divided into multiple geographic regions.Region weights,expressing the regionfunction influence degree on microblog topics,were allocated to regions based on the number of differ-ent POI(Point of Interest)types.Then a sparse additive generative model was applied to generate mi-croblog topic distributions.Beta distributions were employed to depict topic evolution over continuoustime.Finally,we use a Gibbs sampling method to estimate model parameters.Experimental resultsshowed that not only does our model track the temporal distribution of microblog topics but also en-hances topic extraction accuracy when compared with other geography topic models.