|
|
|
| Non-negative matrix factorization model for Web service QoS prediction |
| SU Kai1,2, MA Liang-li2, SUN Yu-fei2, GUO Xiao-ming2 |
| 1.Department of Equipment Economics and Management, Naval University of Engineering, Wuhan 430033, China; 2. Department of Computer Engineering, Naval University of Engineering, Wuhan 430033, China |
|
|
|
Abstract An effective Web service QoS prediction approach was presented by utilizing the neighbor information and latent feature information of the observed QoS data. A matrix factor model for service QoS prediction was presented. Then an expectation-maximization (EM) estimation scenario was designed to learn the model based on the available QoS data. A neighbor information combined non-negative matrix factorization algorithm NCNMF+EM was proposed to implement the scenario. The approach fully utilizes the information of the observed data, and can achieve high prediction accuracy. Experimental results demonstrate that the approach achieves better prediction accuracy than other state-of-the-art approaches. The computational time of the algorithm is linear with the scale of QoS matrix, which indicates that the approach is applicable to large scale QoS prediction problem．
|
|
Published: 10 September 2015
|
|
|
面向Web服务QoS预测的非负矩阵分解模型
针对目前QoS预测算法准确度不高的问题,提出通过挖掘已有QoS观测数据中的近邻信息和隐含特征信息而实现服务QoS预测的方法.建立QoS预测的矩阵分解因子模型,将QoS预测问题转化为稀疏QoS矩阵下的模型参数期望最大化（EM)估计问题,提出结合近邻信息的非负矩阵分解算法NCNMF+EM对该问题进行求解.算法综合利用了QoS矩阵中的近邻信息和隐含特征信息,可以实现对不同类型QoS属性值的准确预测.实验结果表明,采用该方法可以显著地提高服务QoS的预测准确度,且算法的运行时间随着矩阵规模的增大呈线性增长,可以应用于大规模的QoS预测问题中.
|
|
| ［1］ SU K, MA L, GUO X, et al. An efficient discrete invasive weed optimization algorithm for web services selection ［J］. Journal of Software, 2014, 9(3): 709-715．
［2］ MOSER O, ROSENBERG F, DUSTDAR S. Domain- specific service selection for composite services ［J］. IEEE Trans. on Software Engineering, 2012, 38(4): 828-842．
［3］ SU K, MA L, GUO X, et al. An efficient parameter adaptive genetic algorithm for service selection with end-to-end QoS constraints ［J］. Journal of Computational Information Systems, 2014, 10(2): 581-588．
［4］ ZHENG Z, ZHANG Y, LYU R M. Distributed QoS evaluation for real-world web services ［C］∥IEEE International Conference on Web Services. Miami: IEEE, 2010: 83-90．
［5］ YU T, ZHANG Y, LIN K. Efficient algorithms for web services selection with end-to-end QoS constraints ［J］. ACM Transactions on the Web, 2007, 1(1): 126．
［6］ XIAO R. Constructing a novel QoS aggregated model based on KBPP ［J］. Communications in Computer and Information Science, 2010, 107(3): 117-126．
［7］ ALRIFAI M, RISSE T. Combining global optimization with local selection for efficient QoS-aware service composition ［C］∥Proceeding of the 18th International Conference on World Wide Web. New York: ［s.n.］ , 2009: 881-882．
［8］ SHAO L, ZHANG J, WEI Y, et al. Personalized QoS prediction for web service via collaborative filtering ［C］∥IEEE International Conference on Web Services. Salt Lake City: IEEE, 2007: 439-446．
［9］ ZHENG Z, MA H, LYU R M, et al. WSRec: a collaborative filtering based web service recommender system ［C］∥IEEE International Conference on Web Services. Los Angeles: IEEE, 2009: 437-444．
［10］ 刘志中, 王志坚, 周晓峰, 等. 基于事例推理的Web服务QoS动态预测研究［J］. 计算机科学, 2011, 38(2): 119-121．
LIU Zhi-zhong, WANG Zhi-jian, ZHOU Xiao-feng, et al. Dynamic prediction method for web service QoS based on case-based reasoning ［J］. Computer Science, 2011, 38(2): 119-121．
［11］ 张莉, 张斌, 黄利萍, 等. 基于服务调用特征模式的个性化Web服务QoS预测方法［J］. 计算机研究与发展,2013, 50(5): 1070-1071．
ZHANG Li, ZHANG Bin, HUANG Li-ping, et al. A personalized web service quality prediction approach based on invoked feature model ［J］. Journal of Computer Research and Development, 2013, 50(5): 1070-1071．
［12］ ZHENG Z, MA H, LYU R M, et al. Collaborative web service QoS prediction via neighborhood integrated Matrix factorization ［J］. IEEE Transactions on Services Computing, 2013, 6(3): 289-299．
［13］ ZHANG Y, ZHENG Z, LYU R M. WSPred: a time-aware personalized QoS prediction framework for web services ［C］∥IEEE International Symposium on Software Reliability Engineering. Hiroshima: IEEE, 2011: 210-219．
［14］ 彭飞, 邓浩江, 刘磊. 面向个性化服务推荐的QoS动态预测模型［J］. 西安电子科技大学学报：自然科学版, 2013, 40(4): 207-213.
PENG Fei, DENG Hao-jiang, LIU Lei. QoS-aware temporal prediction model for personalized service recommendation ［J］. Journal of Xidian University, 2013, 40(4): 207-213．
［15］ ZHANG S, WANG W, FORD J, et al. Learning from incomplete ratings using non-negative matrix factorization ［C］∥6th SIAM Conference on Data Mining. Seoul: ［s.n.］,2006: 548-552．
［16］ SALAKHUTDINOV R, MNIH A. Probabilistic matrix factorization ［C］∥Proceeding of Advances in Neural Information Processing Systems. Denver: ［s.n.］, 2007: 1257-1264．
［17］ ZHANG S, WANG W, FORD J, et al. Using singular value decomposition approximation for collaborative filtering ［C］∥Proceeding of the 7th IEEE International Conference on E-Commerce Technology. Munich: IEEE, 2005: 18．
［18］ SREBRO N, JAAKKOLA T. Weighted low rank approximation ［C］∥Proceeding of the 20th International Conference on Machine Learning. Washington: ［s.n.］, 2003．
［19］ LEE D D, SEUNG H S. Learning the parts of objects by non-negative matrix factorization ［J］. Nature, 1999, 401(6755): 788-791．
［20］ LEE D D, SEUNG H S. Algorithms for non-negative matrix factorization ［C］∥Proceeding of Advances in Neural Information Processing Systems. Denver: ［s.n.］, 2000: 556-562．
［21］ HU R, PU P. Enhancing collaborative filtering systems with personality information ［C］∥Proceeding of The 5th ACM Conference on Recommender Systems. New York: ［s.n.］, 2011: 197-204. |
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
| |
Shared |
|
|
|
|
| |
Discussed |
|
|
|
|
