Explainable AI-enabled predictive analytics

Overview

Modern predictive analytics underpinned by AI-enabled learning (such as machine learning, deep learning) techniques has become a key enabler to the automation of data-driven decision making. In the context of process monitoring and forecast, predictive analytics has been applied to making predictions about the future state of a running process instance - for example, which task will be carried out next, when and who will perform the task, when will an ongoing process instance complete, what will be the outcome upon completion, etc.

Machine learning models can be trained on event log data recording historical process execution to build the underlying predictive models. Multiple techniques have been proposed so far which encode the information available in an event log and construct input features required to train a predictive model. While accuracy has been a dominant criterion in the choice of various techniques, these techniques are often applied as a black-box in building predictive models.

In this project, we aim to develop new methods and techniques to (machine) learn from event log data to generate accurate and explainable predictions about the future state of ongoing process execution, so that these predictions can be used to provide transparent, trustful and human-understandable insights for decision making.

For more information about this research project, check out the world leading research initiative on eXplainable Analytics for Machine Intelligence (XAMI).

Research activities

The research activities below can be scoped to cater for different types of research student projects:

• design and implementation of models and algorithms for generating process predictions and prediction interpretations (e.g., the reasoning behind predictions)
• design and implementation of methods and techniques for generating user-centric explanations for process predictions
• development and testing of an open-source framework for realising the proposed design and rendering the results to users in an understandable, visual and interactive manner.

Outcomes

• new/improved models and algorithms for generating process predictions and prediction interpretations
• new/improved methods and techniques for generating user-centric explanations for process predictions
• new tools and visualisation of explainable AI-enabled predictive analytics to support human decision-making.

Skills and experience

• knowledge of data mining and machine learning
• programming skills in Python
• knowledge of process mining/analytics (preferable)
• problem-solving and critical thinking skills
• academic writing skills.

Scholarships

You may be eligible to apply for a research scholarship.

Explore our research scholarships

Keywords

• data mining
• machine learning
• deep learning
• predictive analytics
• process analytics

Contact

Contact the supervisor for more information.