Process analytics is a form of data analytics with a focus on extracting data-driven insights from sequences of time-ordered events. Traditionally, process analytics have primarily been used to analyse business processes (e.g., insurance claims handling process, purchase requisition process, and student enrolment process). Nevertheless, the prevalence of timestamped data produced by today’s IT systems has seen a wider application of process analytics to other domains, including industrial control systems (e.g., power plant) and computer network systems, with a goal of enabling a more sophisticated behaviour-based reasoning of faults and security compromises in those systems.
The expansion of process analytics to multiple domains highlight the issues of privacy which has not received sufficient attention within the process analytics community. Process data is likely to contain sensitive information (e.g., employee personal information and sensitive data accessed through a computer network), and insights extracted from process analytics can be privacy-intrusive (e.g., the revelation about sub-optimal performances by certain employees, or personal Internet browsing patterns of a private citizen).
The goal of this research project is to develop tools, techniques, and ultimately, a framework in which the privacy of individuals is preserved to the furthest extent possible by transforming users’ personally identifiable information (typically through cryptographic techniques) such that they are still amenable to data analytics but without compromising the privacy of individuals whose data is being analysed.
A related research in this project is therefore to develop novel process analytics techniques that work with (cryptographically-)transformed data sets.
Research activities in this project vary, depending on students' skills and duration of the project. In general, research acitivites for this project include:
- A literature review of state-of-the-art privacy- preserving data analytics techniques. (VRES)
- Exploration of, and experiments with, cryptographic techniques that either (1) support a balance between individuals’ privacy and the requirements for data analytics, such as homomorphic encryption, or (2) can prevent large-scale, and potentially opportunistic, privacy-intrusive analysis of data involving a massive number of individuals, such as cryptographic proof-of-work techniques. (VRES, Honours, Master)
- The development of new, ‘crypto-friendly’ process analytics algorithms and/or the extension of existing process analytics algorithm to work with a combination of data in the form of both plaintext and ciphertext. (PhD)
- The design and evaluation of an airtight privacy-respecting process analytics framework that provides a certain guarantee of users’ privacy, configurable to suit different threat environment and users’ paranoia appetite. (PhD)
The expected project outcomes are dependent on the scope of the project that students’ undertake. Key outcomes include:
- Gap analysis in the domain of privacy-preserving data analytics techniques
- An evaluation of the readiness of various cryptographic techniques for the purposes of supporting privacy-preserving process analytics (from the perspectives of functionality and performance)
- One or more crypto-friendly process analytics algorithms
- A configurable privacy-respective process analytics framework
Skills and experience
- Familiarity with the fields of data mining, data science, process mining, and/or information security
- Reasonable writing skills
- Problem-solving and logical thinking capabilities
- Computer programming skills
You may be able to apply for a research scholarship in our annual scholarship round.
- data science
- process analytics
- data science
- big data
- information security
- machine learning
Contact the supervisor for more information