As an important means of reading promotion, recommended reading list plays a significant role in cultivating reading habits and enhancing reading interest. The paper proposes a method to generate recommended reading list automatically based on swarm intelligence,aiming to address the problems that book recommendation falls short of user requirements and user acceptance is poor. By comparing the results of this experiment with the book recommendation in Wuhan University Library, it is found that book recommendation based on swarm intelligence is better at meeting the needs of reading promotion, satisfying more user interests, and easier to replicate at scale.