INCREMENTALLY DEFINING ANALYSIS PROCESSES USING SERVICES AND BUSINESS PROCESSES
Increasingly, the idea of using SOA and BPM principles is being applied outside the enterprise computing context. One potential area is the use of reusable services to facilitate the definition, composition and execution of analysis processes over large distributed repositories of heterogeneous data. However, it is hard to see how such a solution can be applied when users are already engaged in performing the same tasks using their own tools and processes. This paper studies the problem in more detail and proposes a simple method in which the initial analysis process is iteratively refined into a service-based one. The methodology relies on an ADAGE architecture which gives the flexibility to define data analysis processes in an incremental way. As a case study, the approach is demonstrated on the process of conducting event studies using the SAS package together with different data sources (e.g. Thomson Reuters Tick History-- TRTH). The case study demonstrates that the resulting process is substantially more efficient and effective than the original one.
Tick data Data-intensive analysis Event study Web services Business process SOA
Weisi Chen Fethi A. Rabhi
School of Computer Science and Engineering, University of New South Wales, Sydney 2052, Australia
国际会议
13th International Conference on Enterprise Information System(第13届企业信息系统国际会议 ICEIS 2011)
北京
英文
1070-1073
2011-06-08（万方平台首次上网日期，不代表论文的发表时间）