«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

30

期数:

2020年03期

页码:

93-97

栏目:

智能、算法、系统工程

出版日期:

2020-03-10

文章信息/Info

Title:

An Adaptive Web Service QoS Prediction Method Based on Collaborative Filtering

文章编号:

1673-629X(2020)03-0093-05

作者:

庄 崟¹; 郭志川²; 3; 黄逍颖²; 3

1.江苏有线技术研究院有限公司,江苏 南京 210001; 2.中国科学院声学研究所国家网络新媒体工程技术研究中心,北京 100190; 3.中国科学院大学,北京 100049

Author(s):

ZHUANG Yin¹; GUO Zhi-chuan²; 3; HUANG Xiao-ying²; 3

1.Jiangsu Cable Technology Research Institute Co.,Ltd.,Nanjing 210001,China; 2.National Network New Media Engineering Research Center,Institute of Acoustics, Chinese Academy of Sciences,Beijing 100190,China; 3.University of Chinese Academy of Sciences?Beijing 100049,China

关键词:

协同过滤; Web服务; QoS预测; 自适应; 客户端

Keywords:

collaborative filtering; Web service; QoS prediction; adaptive; client

分类号:

TP317

DOI:

10. 3969 / j. issn. 1673-629X. 2020. 03. 018

摘要:

随着Web服务越来越多,服务质量 QoS 作为描述Web服务的非功能性属性变得越来越重要。 通常,一种服务的 QoS 对用户来说是未知的,因此对于基于Web服务的应用,精确预测其未知的QoS对于成功部署该服务具有重要的价值。 基于协同过滤的WSRec算法是一种高精度的QoS预测方法,为进一步提升QoS的预测精度,提出了一种协同过滤的自适 应Web服务QoS预测方法。 该方法通过客户端首先发出QoS-Web服务请求;服务端接到请求后,根据已有数据,计算两 两用户或服务间的相似度;并根据相似性找到对于目标用户的 K 个最接近用户或服务,生成该QoS值预测值 A;同时在计 算相似性时,采用改进皮尔逊相关系数得到预测值 B;最后将预测值 A 和 B 以权值相结合得到目标用户或服务的QoS值。 该算法改进了单一的协同过滤在数据稀疏的情况下,对相似性给予过高估计的不足,使得QoS预测值精度得以提高,取得了更好的实验结果。 实验表明该方法预测精度优于WSRec算法。

Abstract:

With more and more Web services,quality-of-service(QoS) is becoming more and more important for as a non-functional attribute describing Web services. Generally,the QoS values of a service are unknown to its users,so the accurate prediction of unknown QoS values is significant for the successful deployment of Web service-based applications. WSRec algorithm based on collaborative filtering is a highly accurate method for predicting QoS. In order to improve the accuracy of QoS prediction further,an adaptive Web service QoS prediction method based on collaborative filtering is proposed. This method firstly sends QoS-based Web request to the server through the client. After receiving the request,the server calculates the similarity between each of the two users or between each of the two services based on the QoS data. At the same time,according to these similarities,theKclosest users or services to the target user are found,and the predicted valueAof QoS is generated. When calculating the similarity,the predicted Pearson correlation coefficient is used to obtain the predicted value B. Finally,the predicted values A and Bare given to obtain the QoS value by changing the weight between the two values.The algorithm improves the accuracy of QoS prediction by improving the shortcomings of a single collaborative filtering to overestimate the similarity in the case of sparsely populated data,and obtains a better experimental result.The experiment shows that the proposed method achieves better prediction accuracy than WSRec algorithm.

常用功能

更新日期/Last Update: 2020-03-10