Process mining is a specialised form of analytics of growing practical significance. It is well known for its ability to mine datasets, resulting from the execution of processes, to identify inefficiencies, and to reveal insights into resource behaviour, activity dependencies, and process compliance, among others.
As such, it is instrumental to effective decision making and streamlining of business processes. Process mining can also operate in tandem with process automation to fine tune on-the-fly decision making and predicting resource utilisation.
A challenge that remains under-researched is to allow process mining approaches to scale up and be able to deal with very large data sets. To this end, technologies from the field of databases could be exploited. For example, the map-reduce paradigm aims to parallelise computations in order to speed up complex time-task consuming tasks.
This paradigm has been implemented in Hadoop. The central question of this project is to investigate to what extent and how the map-reduce approach can help process mining become more scalable.
You will work with your supervisor and project coordinator in an area of potential contribution to theory and practice. This project involves the following:
- A literature review of the current implementation of map-reduce approach in the process mining domain
- Development of use case scenarios demonstrating the application of map-reduce approach for process mining problems
- Suitability of usage of map-reduce approach for process mining problems by using real life event log
- An understanding of the map-reduce approach and its application in the process mining domain,
- An implemented map-reduce approach for process mining.
Skills and experience
The student is expected to have sound technical understanding of business process management and databases.
You may be able to apply for a research scholarship in our annual scholarship round.
Contact the supervisor for more information