针对协同过滤算法数据稀疏性和冷启动的问题,提出一种基于内容相似度的知识协同过滤推送算法。该算法在原有协同过滤算法的基础上,综合考虑了用户兴趣和项目属性的影响,利用内容相似度的方法解决了用户一项目评分矩阵的数据稀疏性等问题。该算法有效地提高了知识推送的精确度,满足了用户的知识需求。
As for the data sparsity and cold start problems in the collaborative filtering algorithm, this paper proposes a knowledge collaborative filtering and push algorithm based on content similarity. Based on the original collaborative filtering algorithm, this algorithm takes the influences of user interest and project attribute into full account, and uses the content similarity method to solve the data sparsity problem in the user-project scoring matrix. The algorithm improves the accuracy of knowledge push effectively, and meets the user demand for knowledge.