基于协同过滤和隐语义模型的混合推荐算法

协同过滤算法一般根据用户的评价信息来推测用户的喜好,但受到数据稀疏问题的影响,很多时候无法得到较为理想的推荐结果;除此之外,一般协同推荐算法忽略了用户兴趣的动态变化;文中提出的算法主要融合了相似度传递、用户兴趣迁移、隐语义模型等用以解决上述问题。首先提出了基于项目相似度的协同推荐算法。该算法深入研究了改进的余弦相似度方法,在执行过程中首先需要对项目进行信任关系建模,基于此来传递相似度,然后将这两部分相似度关系进行加权得到新的项目相似关系,可以将其应用到项目的评分中。其次,提出了基于用户兴趣迁移的隐语义模型推荐算法。该算法引入时间函数,重构用户的兴趣模型,实现对传统模型的修正,然后再使用梯度下降法来求解。最后,采用线性融合的办法,将以上两种算法进行融合。实验对比结果表明,混合推荐算法的推荐准确率较原先的算法有了较大的提高,因为它可以对丢失的信息进行补充,对于用户兴趣的变化能够较好的适应,同时大大弱化了数据的稀疏导致的一系列负面影响。

A Hybrid Recommendation Algorithm Based on Collaborative Filtering and Latent Factor Model

The collaborative filtering algorithm generally estimates the user’s preferences based on their evaluation information.However,due to the data sparsity,in many cases,it is impossible to obtain the ideal recommendation results.In addition,the collaborative recommendation algorithm ignores the dynamic change of user interest.The algorithm proposed in this paper mainly combines similarity transfer,user interest migration and latent factor model to solve the above problems.Firstly,we propose a collaborative recommendation algorithm based on project similarity transfer,which improves the cosine similarity method.In this algorithm,we model the trust relationship of the project and transfer the similarity based on it,then we weigh the similarity relationship between the two parts to get a new project similarity relationship,which can be applied to the project scoring.Secondly,we propose a latent semantic model recommendation algorithm based on user interest migration.The algorithm introduces the time function,reconstructs the user’s interest model to implement the modification of the traditional model,and then uses the gradient descent method to solve.Finally,we use the linear fusion method to fuse the above two improved algorithms.The experiment shows that the linear fusion method is better because it can supplement the missing information,adapt to the change of user interest,and greatly weaken a series of negative effects caused by sparse data.