主页期刊介绍编委会编辑部服务介绍道德声明在线审稿编委办公编辑办公English
     
在线出版
各期目录
纸质出版
分辑系列
论文检索
论文排行
综述文章
专刊文章
美文分享
各期封面
E-mail Alerts
RSS
旧版入口
中国科学院软件研究所
  
投稿指南 问题解答 下载区 收费标准 在线投稿
庄媛,张鹏程,李雯睿,冯钧,朱跃龙.一种环境因素敏感的Web Service QoS监控方法.软件学报,2016,27(8):1978-1992
一种环境因素敏感的Web Service QoS监控方法
Web Service QoS Monitoring Approach Sensing to Environmental Factors
投稿时间:2014-11-28  修订日期:2015-04-22
DOI:10.13328/j.cnki.jos.004850
中文关键词:  服务质量  影响因子  TF-IDF算法  加权朴素贝叶斯分类器  监控
英文关键词:quality of service  impact factor  TF-IDF algorithm  weighted naïve Bayesian classifier  monitor
基金项目:国家自然科学基金(61572171,61202097,61202136,61370091);高等学校博士学科点专项科研基金(20120094120009);江苏省自然科学基金(BK20130852);中央高校基本科研业务费(B15020191)
作者单位E-mail
庄媛 河海大学 计算机与信息学院, 江苏 南京 211100  
张鹏程 河海大学 计算机与信息学院, 江苏 南京 211100 pchzhang@hhu.edu.com 
李雯睿 可信云计算与大数据分析重点实验室(南京晓庄学院), 江苏 南京 211171  
冯钧 河海大学 计算机与信息学院, 江苏 南京 211100  
朱跃龙 河海大学 计算机与信息学院, 江苏 南京 211100  
摘要点击次数: 2520
全文下载次数: 1510
中文摘要:
      面向服务系统的执行能力依赖第三方提供的服务,在复杂多变的网络环境中,这种依赖会带来服务质量(QoS)的不确定性.而QoS是衡量第三方服务质量的重要标准,因此,有效监控QoS是对Web服务实现质量控制的必要过程.现有监控方法都未考虑环境因素的影响,比如服务器位置、用户使用服务的位置和使用时间段负载等,而这些影响在实际监控中是存在的,忽略环境因素会导致监控结果与实际结果有悖.针对这一问题,提出了一种基于加权朴素贝叶斯算法wBSRM(weightednaive Bayes running monitoring)的Web Service QoS监控方法.受机器学习分类方法的启发,通过TF-IDF(term frequency-inverse document frequency)算法计算环境因素的影响,通过对部分样本进行学习,构建加权朴素贝叶斯分类器.将监控结果分类,满足QoS标准为c0,不满足QoS标准为c1,监控时调用分类器得到c0c1的后验概率之比,对比值进行分析,可得监控结果满足QoS属性标准、不满足QoS属性标准和不能判断这3种情况.在网络开源数据以及随机数据集上的实验结果表明:利用TF-IDF算法能够准确地估算环境因子权值,通过加权朴素贝叶斯分类器,能够更好地监控QoS,效率显著优于现有方法.
英文摘要:
      The execution capacity of service-oriented system relies on the third-party services. However, such reliance would result in uncertainties in consideration of the complex and changeable network environment. Hence, runtime monitoring technique is required for service-oriented system. Effective monitoring technique towards Web QoS, which is an important measure of third-party service quality, is necessary to ensure quality control on Web service. Several monitoring approaches have been proposed, however none of them consider the influences of environment including the position of server and user usage, and the load at runtime. Ignoring these influences, which exist among the real-time monitoring process, may cause monitoring approaches to produce wrong results. To solve this problem, this paper proposes a new environment sensitive Web QoS monitoring approach, called wBSRM (weighted Bayes runtime monitoring), based on weighted naive Bayes and TF-IDF (Term Frequency-Inverse Document Frequency). The proposed approach is inspired by machine learning classification algorithm, and measures influence of environment factor by TF-IDF algorithm. It constructs weighted naïve Bayes classifier by learning part of samples to classify monitoring results. The results that meet QoS standard are classified as c0, and those that do not meet is classified as c1. Classifier can output ratio between posterior probability of c0 and c1, and the analysis can lead to three monitoring results including c0, c1 or inconclusive. Experiments are conducted based on both public network data set and randomly generated data set. The results demonstrate that this approach is better than previous approaches by accurately calculating environment factor weight with TF-IDF algorithm and weighted naïve Bayes classifier.
HTML  下载PDF全文  查看/发表评论  下载PDF阅读器
 

京公网安备 11040202500064号

主办单位:中国科学院软件研究所 中国计算机学会 京ICP备05046678号-4
编辑部电话:+86-10-62562563 E-mail: jos@iscas.ac.cn
Copyright 中国科学院软件研究所《软件学报》版权所有 All Rights Reserved
本刊全文数据库版权所有,未经许可,不得转载,本刊保留追究法律责任的权利