提出了一种运用模糊聚类方法将项目属性特征的相似性与协同过滤推荐算法相融合的推荐方法,此方法将用户对单个项目的偏好转化为对相似群组的偏好,目的是构造密集的用户-模糊簇的偏好信息,同时利用项目之间在相似群组的相似性来初步预测用户对未评价项目的评分,在此基础之上再完成基于用户的协同过滤推荐算法。实验结果表明,该方法确实可提高协同过滤推荐算法的推荐精度。 This paper proposes a method of fuzzy clustering to combine the similarity of project attribute features with the collaborative filtering recommendation algorithm. This method converts the user’s preference of a single item into the preference of similar groups in order to construct a dense User-fuzzy cluster preference information, and at the same time, the similarity of similar groups in the project is used to predict the user’s score of the non-evaluated project. Based on this, the user-based collaborative filtering recommendation algorithm is completed. The experimental results show that this method can indeed improve the recommendation accuracy of the collaborative filtering recommendation algorithm.