Research on Book Recommendation Algorithm for Library Readers Based on MMR-DeepFM

为更为有效地提高图书馆图书资源使用率并为读者提供个性化推荐服务，本文提出了一种融合最大边缘相关(Maximal-Marginal-Relevance, MMR)算法和深度因子分解机(Deep Factorization Machine, DeepFM)深度学习神经网络的MMR-DeepFM图书推荐模型。采用中国海洋大学信息学部读者数据进行了模型训练和测试，与其它图书推荐算法比较，本文提出的图书推荐算法可以更为准确和多样化地为读者推荐图书，最终提高图书馆图书的借阅率。To enhance the utilization rate of library book resources and provide personalized recommendation services for readers more effectively, an MMR-DeepFM book recommendation model that integrates the Maximal-Marginal-Relevance (MMR) algorithm and the Deep Factorization Machine (DeepFM) deep learning neural network is proposed in this paper. The model was trained and tested using reader data from the Information Department of Ocean University of China. Compared with other book recommendation algorithms, the proposed algorithm in this paper can recommend books more accurately and diversely to readers, ultimately increasing the borrowing rate of library books.