The invention discloses a recommendation method considering an overlapping community effect, and the recommendation method is characterized by including the following steps of 1, obtaining a user-theme probability distribution matrix and a theme-commodity probability distribution matrix; 2, calculating the utility of the commodity for the user; and 3, according to the ranking of the utility scorefrom high to low, using the top N items to generate a recommendation list for the user and recommending the recommendation list to the user. The method can effectively solve the problem of 'cold start' and data sparsity in the existing recommendation system so as to improve the recommendation quality and provide better personalized service for the user.