Thesis Topic Details

Topic ID:
3388
Title:
Efficient mining of changing business processes
Supervisor:
Raymond Wong
Research Area:
Business Process Management, Services, Software Engineering
Associated Staff
Assessor:
Helen Hye-Young Paik
Topic Details
Status:
Active
Type:
R & D
Programs:
CS CE BIOM BINF SE
Group Suitable:
Yes
Industrial:
No
Pre-requisites:
Good understanding of Web services or knowledge on process management is a plus.
Description:
Process mining is a process management technique analyses business processes based on event logs. The basic idea is to extract knowledge from event logs recorded by an information system. This knowledge, in the form of business process models, can be extracted by process mining algorithms. These mined models can be used for the purpose of process compliance, and improving current business models. While many process mining algorithms have been proposed recently, there does not exist a widely-accepted benchmark to evaluate and compare these process mining algorithms. As a result, it can be difficult to choose a suitable process mining algorithm for a given enterprise or application domain.

For instance, process mining algorithms are used to mine business process models using process logs. The mined models will then be compared against the original process models of the enterprise for conformance checking, or to discover more efficient, streamlined business process models (which are still semantically equivalent to the original process models). However, since different process mining algorithms have different properties (e.g., some perform better on models with invisible tasks, while others do not), being able to select the most appropriate process mining algorithm (i.e. the one that produces mined models that are semantically similar to the original models and structurally equal to or better than the original models) for the models from a given enterprise is a big advantage.

While recent research and software prototypes have attempted to provide such an evaluation framework, empirically evaluating all available process mining algorithms against the business models provided by a given enterprise is usually computationally expensive and time consuming.

Worst still, business process models are usually evolving. Therefore, these process model mining algorithms will need to be regularly re-evaluated against these changing models for conformance checking, re-engineering or discovery of more streamlined, improved models.

It is the primary aim of this project to investigate a scalable solution that can evaluate, compare and rank these process mining algorithms efficiently. In particular, it will extend our previous work on choosing an effective process mining algorithm for an enterprise without evaluating different process mining algorithms extensively, and also how to avoid re-evaluating all the algorithms whenever the business processes of the enterprise change.

Report on the work done so far can be found at:
http://www.cse.unsw.edu.au/~wong/papers/tsc12.pdf

Further information of this project can be obtained by contacting the supervisor directly.
Comments:
--
Past Student Reports
  Qing XU in s1, 2014
Efficient mining of changing business processes
 

Download report from the CSE Thesis Report Library

NOTE: only current CSE students can login to view and select reports to download.