融合灰关联分析和信任度的Slope One算法研究

Research on Slope One Algorithm Integrating Grey Association Analysis and Trust

作为一种有效的协同过滤算法,Slope One算法被广泛地应用到一些推荐系统中.但该算法在评分预测时仅根据项目之间的评分偏差以及用户的历史评分信息,未能考虑用户之间的潜在联系,而且数据稀疏性和冷启动问题也会限制算法的推荐性能.因此,为了提高算法的推荐质量,本文采用均衡接近度灰关联方法度量用户关联度,从而有效挖掘用户之间的潜在联系;并引入用户信任模型来缓解冷启动问题对算法的影响,将均衡接近度和信任度作为权重因子加权到评分预测过程.论文研究并提出了一种融合灰关联分析和信任度的Slope One算法,并通过实验验证了算法的有效性.实验结果表明,该算法可以有效地提高算法预测准确度,缓解数据稀疏性和冷启动问题对算法的影响.

As an effective collaborative filtering algorithm,Slope One algorithm has been widely applied to some recommendation systems.However,when predicting scores,the algorithm only relied on the score difference between items and the user’s historical score information,which fails to consider the potential connection between users.Moreover,this algorithm was greatly affected by the data sparsity,which further affected the recommendation quality of the algorithm.Therefore,in order to improve the recommendation effect of the algorithm,this paper introduces the grey correlational analysis by the method of degree of balance approach to measure user relationships,in this way,the potential connection between users can be effectively mined.And introduce user trust model to effectively alleviate the impact of the cold start problem.The degree of balance and approach and trust are weighted as weighting factors to the predictive scoring process.This paper studies and proposes a Slope One algorithm which integrates grey correlation analysis and trust,and verifies the effectiveness of the algorithm through experiments.Experimental results show that the proposed algorithm can effectively improve the prediction accuracy,and alleviate the impact of data sparsity and cold start problems on the algorithm.