论文部分内容阅读
针对Web服务推荐现有技术缺乏显式打分数据缺点,提出使用隐反馈知识进行推荐的方法.该方法首先构造一个伪评分生成器,将用户隐反馈知识映射成为显式打分.基于矩阵因子分解模型,将信任知识引入服务推荐过程,建立一种融合社交信任信息的服务推荐模型,有效提高了服务推荐性能.实验表明,本文提出的基于隐反馈的服务推荐方法预测性能优于最近邻方法和SVD++方法;同SVD++方法的性能对比实验表明,引入信任知识能够进一步提高服务推荐的性能,具有较好的实际应用价值.
Aiming at the disadvantage that the existing technology of web service lacks explicit scoring data, a method of recommending using hidden feedback knowledge is proposed.This method first constructs a pseudo-scoring generator and maps the user’s implicit feedback knowledge into explicit scoring. Based on the matrix factorization model , Introduces the trust knowledge into the service recommendation process and establishes a service recommendation model that integrates the social trust information to effectively improve the service recommendation performance.The experimental results show that the proposed service recommendation method based on implicit feedback is better than the nearest neighbor method and SVD ++ Method; the performance comparison with the SVD ++ method shows that the introduction of trust knowledge can further improve the performance of service recommendation, and has good practical application value.