基于项目的非邻近序列模式推荐算法

Item-based Non-neighbouring Sequential Pattern Recommendation Algorithms

针对推荐系统存在的稀疏性问题,提出将非邻近序列模式挖掘算法与基于项目的协作过滤推荐算法相结合的推荐方法,通过构造Markov概率的路径加权转移矩阵,计算资源被推荐的可能性,向用户进行推荐。结果证明,在数据稀疏的情况下,较传统的基于项目的协作过滤推荐算法,该算法能有效提高推荐系统的推荐质量。

To solve the sparsity problems of recommender systems, a recommendation method is designed by means of the combination of the non-neighbouring sequential pattern mining algorithms and the item-based Collaborative Filtering（CF） recommendation algorithms. By constructing the Markov probability transfer matrix according to the paths weight sum algorithms, it computes the recommendation possibility of resources and recommends to users. Experimental results show that, on the condition of sparse data, it can improve the recommendation quality compared with the traditional item-based Collaborative Filtering recommendation algorithms.