The invention relates to the technical field of personalized recommendation, in particular to a personalized recommendation method fusing social information. In accordance with the user-item scoring matrix, calculating scoring similarity between users, and screening the nearest neighbor set of scoring; then, according to the nearest neighbor set, calculating the prediction score of the target user; calculating the social similarity between users according to the social network information of users, and screening the social nearest neighbor set; according to the social nearest neighbor set, calculating the prediction score of the target user; by combining the two, predicting the target user 's rating value of the project, and descending the order, recommending the K items with the highest rating to the target user, and generating the recommendation list. Finally, the experimental results show that the performance of the proposed personalized recommendation method is better than the current recommendation method, which can effectively improve the accuracy of recommendation, so as to alleviate the data sparseness and cold start problems.