节点文献

基于消息日志的Web服务接口业务协议挖掘

Web Service Business Protocol Mining Based on Message Logs

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 李翔怀进鹏刘旭东孙海龙曲先洋

【Author】 LI Xiang,HUAI Jin-Peng,LIU Xu-Dong,SUN Hai-Long,QU Xian-Yang(School of Computer Science and Engineering,BeiHang University,Beijing 100191,China)

【机构】 北京航空航天大学计算机学院

【摘要】 Web服务接口的业务协议描述了Web服务的外部行为,对于Web服务的复用具有重要意义,可以作为服务发现、组合、验证和运行期可信保障等方面的重要基础.目前,已有一些工作研究了Web服务的协议发现问题,即从Web服务的调用消息日志中挖掘Web服务接口的业务协议.但已有方法主要关注服务的控制流约束,忽略了数据流约束以及数据流和控制流的相互约束.针对这一问题,研究了如何从Web服务的调用日志中自动挖掘Web服务接口,并侧重综合考虑Web服务的数据流和控制流.首先扩展了传统Petri网,提出了一种增加了数据流描述的Web服务接口模型——BPN(business protocol net)模型.在此基础上,进一步提出了一种自动化的挖掘框架,可以从Web服务调用消息记录中合成Web服务的BPN表示.最后,通过仿真实验验证了该方法的有效性.其结果表明,所提出的挖掘算法是正确而有效的.

【Abstract】 A Web service business protocol is used to describe the external behavior of a service and plays an important role in the service discovery,composition,verification,runtime service trustworthy guarantee,and so on. Presently,some research has been done on discovering the Web service business protocol from the invocation logs. Most of these works focused on the control-flow of Web service business protocols that give a temporal constraint among the operations of Web service.However,the data constraints and the consistency between the data-flow and the control-flow are also important and have not received enough attention.This paper studies the Web service business protocol from the service invocation logs and focuses on mining the relations,or the constraints between the message values and service operations.This paper proposes a Petri-net based model,called Business Protocol Net(simply,BPN),to describe the behavior of a service.Based on this model,a mining framework is proposed to automatically generate the BPN model from message traces.Experimental results illustrate that the method is effective in discovering the Web service business protocol from invocation logs.

【基金】 国家高技术研究发展计划(863)(2007AA010301,2006AA01A106,2009AA01Z419)
  • 【文献出处】 软件学报 ,Journal of Software , 编辑部邮箱 ,2011年07期
  • 【分类号】TP393.09
  • 【被引频次】6
  • 【下载频次】359
节点文献中: 

本文链接的文献网络图示:

本文的引文网络