基于动态迭代采样的异质信息网络推荐算法

Research on Heterogeneous Information Network Recommendation Algorithm Based on Dynamic Iterative Sampling

随着数字信息时代的来临,推荐系统以及相应的数据推荐算法承担着越来越重要的角色。面对信息过载的压力下,如何解决传统数据推荐算法存在的信息稀疏问题,同时更好推荐对应的数据信息,表达出用户偏好信息,提出了基于动态迭代采样的异质信息网络推荐算法。上述算法在异质信息网络推荐算法基础上引入动态迭代采样,建立动态数据推荐模型。上述模型通过对不同网络节点信息进行迭代采样,反馈优化各网络节点信息,从而实现高质量信息的推荐。最后,不同的仿真结果表明,所提算法在解决信息稀疏以及复杂任务推荐方面有较好的有效性。

With the advent of the digital information age,recommendation system and corresponding data recommendation algorithms play a more and more important role.Facing the pressure of information overload,how to solve the problem of information sparsity in the traditional data recommendation algorithm,meanwhile,better recommend the corresponding data information and express the user preference information,a heterogeneous information network recommendation algorithm based on dynamic iterative sampling is proposed.Under the pressure of information overload,in order to solve the problem of information sparsity in traditional data recommendation algorithms,better recommend the corresponding data information recommended,and express user preference information,this paper proposes a heterogeneous information network recommendation algorithm based on dynamic iterative sampling.Based on the heterogeneous information network recommendation algorithm,the we introduced dynamic iterative sampling and established a dynamic data recommendation model.By iteratively sampling the information of different network nodes,the information of each network node was fed back and optimized to realize the recommendation of high-quality information.Finally,different simulation experiments show that the algorithm is effective in solving sparse information and complex task recommendation。