基于作者的协同过滤图书推荐算法

Collaborative Filtering Book Recommendation Algorithm Based on Author

选择准确的图书推荐算法,不仅可以让读者快速而有效地获得所需图书信息,减少检索搜寻时间,同时也可以较好地发挥图书馆馆藏资源的潜在价值。在协同过滤推荐算法的理论与方法的基础上,针对目前比较常用的两种协同过滤图书推荐算法的不足,从作者与作者的相似性角度出发,提出了一种基于作者的协同过滤图书推荐算法。该方法包括计算作者的特征向量、计算相似度、逻辑打分、TOP-K推荐等步骤。算法示例给出了该算法的基本原理与实现过程,结果表明该算法简便、可行,为图书管理推荐工作提供了一种新的思路。

Choosing accurate book recommendation algorithm not only allows readers to obtain the required book information quickly and efficiently,reduces the retrieval and search time,but also gives better play to the latent value of library collection resources.Based on the theory and method of collaborative filtering recommendation algorithm,in view of the shortcomings of the two commonly used collaborative filtering book recommendation algorithms,from the perspective of the similarity between the authors,collaborative filtering book recommendation algorithm based on authors is proposed.The method includes calculating the author's feature vector,calculating similarity,logical scoring,TOP-K recommendation and other steps.The algorithm example shows the basic principle and implementation process of the algorithm,and the results show that the algorithm is simple,convenient and feasible,which provides a new idea for book management recommendation work.