Queen's University - Utility Bar

QSpace at Queen's University >
Graduate Theses, Dissertations and Projects >
Queen's Graduate Theses and Dissertations >

Please use this identifier to cite or link to this item: http://hdl.handle.net/1974/1564

Authors: Guo, JIN

Files in This Item:

File Description SizeFormat
Guo_Jin_200810_MSc.pdf1.56 MBAdobe PDFView/Open
Keywords: buisness processes
clone detection
business applications
UI design patterns
Issue Date: 2008
Series/Report no.: Canadian theses
Abstract: A business application automates a collection of business processes. A business process describes how a set of logically related tasks are executed, ordered and managed by following business rules to achieve business objectives. An “online book purchase” business process contains several tasks such as buying a book, ordering a book, and sending out promotions. In this ever changing business environment, both of business applications and business processes are modified to accommodate changed business requirements and improve the performance of the organization. These continuous modifications introduce problems in the following two aspects: 1) Business process definitions are rarely updated to reflect the current business processes deployed in business applications. 2) Business processes may be cloned (e.g., copied and slightly modified) to handle special circumstances or promotions. Identifying these clones and removing them help improve the efficiency of an organization. However, business processes are defined with textual languages that cannot be automatically understood. To maintain business process definitions up to date, we present our techniques that automatically recover business processes from UIs of business applications and identify clones in the recovered business processes. We leverage UI design patterns, which present the best practices of UI designs, to capture business processes from UIs. To refine the recovered business processes and mark the functionally equivalent tasks, we use existing code clone detection tools, such as CCFinder and CloneDR, to detect clones in business applications, and lift clones from code level to business process level. The effectiveness of our techniques is demonstrated through a case study on 15 large open source business applications.
Description: Thesis (Master, Computing) -- Queen's University, 2008-10-28 11:06:31.41
URI: http://hdl.handle.net/1974/1564
Appears in Collections:School of Computing Graduate Theses
Queen's Graduate Theses and Dissertations

Items in QSpace are protected by copyright, with all rights reserved, unless otherwise indicated.


  DSpace Software Copyright © 2002-2008  The DSpace Foundation - TOP