Slope One algorithm based on a simple linear regression model. Reducing the response time and mainte-nance difficulty, it significantly improve the recommended performance. However, Slope One algorithm does not con-sider the internal relevance of users. Using data of all users without distinction is likely to cause deviation and effect the recommendation quality. In this paper we propose an improved Slope One algorithm which takes user similarity into account and modifies the rating deviation calculation formula. Combing item-based Slope One algorithm and user-based collaborative filtering algorithm, a new hybrid recommendation algorithm US-Slope One is proposed. The experimental results on Movielens data set show that the proposed algorithm has better prediction accuracy and rec-ommendation quality compared with the original Slope One algorithms.%Slope One算法基于简单的线性回归模型，通过减少响应时间和维护难度，显著提高了推荐性能。然而Slope One算法没有考虑用户内部的关联，同等地使用各个用户数据进行预测，容易造成偏差，影响推荐质量。本文提出了一种改进的Slope One算法，它将用户相似度纳入考虑并且对评分偏差计算公式进行了修正。基于项目的Slope One算法结合基于用户的协同过滤算法，提出新的混合推荐算法US-Slope One。在MovieLens数据集上的实验结果表明，该算法与原Slope One算法相比具有更好的预测准确度和推荐质量。
展开▼
