中文摘要：

本文提出了一种融入个体人格特质的个性化图书推荐算法RecUBSPre，该算法基于个体背景因素漂移准确定位目标数据集；并针对图书评分数据的稀疏性，引入用户图书类型偏好矩阵，得出用户图书偏好类型；最后利用“大五”人格模型，对人格特质兼容度进行定义，消除了最近邻用户集合中由于用户盲从心理导致的评分有失偏颇现象，从而优化图书推荐集的排序，提高用户满意度。实验结果表明，该算法能提高图书推荐的准确度。

英文摘要：

This paper puts forward a personalized book recommendation algorithm integrated with individual personality traits called RecUBSPre. The algorithm consider the drift of the individual background factors set, so as to accurately locate the target data set. And in order to solve the data sparseness of books score data, introduction of user preference matrix book type, so it is concluded that the user preference type books. Finally, using the ＂ big five＂ personality model, the personality compatibility degree is defined, eliminates the nearest neighbor users set score as biased due to user blindly follow the psychological phenomenon, thus books recommended set of optimized scheduling, improve customer satisfaction. The experimental results show that the algorithm can improve the accuracy of book recommendation.