融合初始资源与协同过滤的二部图推荐算法

推荐系统的产生主要是为了解决信息过载的问题。基于二部图网络与基于协同过滤的推荐算法是目前应用比较广泛的算法,二者都取得了一定的推荐效果。基于加权二部图网络的算法忽略对初始资源的配置,基于物品的协同过滤算法在推荐时也产生数据稀疏等问题。组合推荐算法融合初始资源配置以及基于物品的协同过滤算法来解决相关的问题,可以达到更好的推荐效果。算法实验在MovieLens数据集上实施,结果表明,与传统的推荐算法以及最近的组合推荐算法相比,该方法有更好的推荐效果。

BIPARTITE GRAPH RECOMMENDATION ALGORITHM BASED ON INITIAL RESOURCES AND COLLABORATIVE FILTERING

The recommendation system aims to solve the problem of information overload.The recommendation algorithms based on bipartite graph network and collaborative filtering are widely used at present.Both of them have achieved certain recommendation results.The algorithm based on the weighted bipartite graph network ignores the configuration of initial resources,and the item-based collaborative filtering algorithm has problems of data sparsity and other issues.The combined recommendation algorithm which combined initial resource configuration with item-based collaborative filtering algorithm could achieve a better recommendation effect to solve the problems.The experiment was implemented on the MovieLens data set.The results show that the method achieves a better recommendation effect compared with the traditional and current combined recommendation algorithms.