论文部分内容阅读
在为用户选择并组合其满意的Web服务时,预测Web服务缺失的服务质量(quality of service,QoS)值是必要的。为解决该问题,提出一种R-SRec模型,加入用户地理位置信息和服务的反向预测,提高预测精准度。在基于用户的协同过滤算法中融入用户地理位置信息,提高参与预测数据的空间相关度;在基于服务的协同过滤算法中加入Web服务反向预测,缓解数据稀疏问题;根据不同的置信度融合两种算法,对缺失的QoS值进行预测。使用真实的数据集与其它3类常用的算法进行比较,实验结果表明,该方法的预测结果精确度更高。
Predicting the missing quality of service (QoS) value for Web services is necessary when choosing and combining the user’s favorite Web services. In order to solve this problem, an R-SRec model is proposed, which adds the reverse prediction of users’ geographic location information and services to improve the prediction accuracy. In the service-based collaborative filtering algorithm, Web service reverse prediction is added to alleviate the problem of data sparseness. Based on the fusion of two confidence levels, the spatial location of the users is improved. Algorithm to predict missing QoS values. The real data set is compared with the other three commonly used algorithms. Experimental results show that the proposed method has higher accuracy.