Description

Despite the increasing adoption of Explainable AI (XAI) in Learning Analytics (LA), many
Explainable Learning Analytics (XLA) systems used to support competence-based
assessment (CBA) frameworks in education remain non-interactive. While these systems
can quantify learner performance across competencies, existing dashboards/platforms
tend to present static visualizations and summary-level metrics, offering little room for
learner interaction, scenario exploration, or personalized insight extraction. Without
interactivity, explanations remain surface-level and fail to accommodate individual
learning paths and cognitive diversity. Therefore, there is a critical need for a new
generation of interactive, explainable learning analytics tools that provide real-time,
learner-specific justifications for the analytics, allow users to query, explore, and reflect on
their assessment data; support personalized pathways and interventions based on
transparent insights; and uphold fairness, trust, and accountability in the evaluation
process. Addressing this gap is essential for aligning learning analytics with the core
values of competence-based education, i.e., clarity, mastery, feedback, and learner
autonomy.

This will be an implementation-focused study, blending system engineering, human-
computer interaction, educational theory, and responsible AI principles.

Main Objective:

To develop and evaluate an Interactive Explainable Learning Analytics (InXLA) platform
that enhances transparency, engagement, personalisation and actionable feedback in a
competence-based assessment digital environment.

Specific Objectives:

1.  To identify the requirements and design principles for implementing InXLA within
   competence-based assessment systems.
2. To develop an InXLA framework that integrates explainable AI techniques with
   interactive learner and educator interfaces
3. To develop a functional prototype (dashboard) that provides learners and educators:
   • Real-time visualizations of competence levels and progress
   • Underlying explanatory factors or justifications for the analytics
4. To integrate learner- and educator-driven interactivity, enabling both groups to query
   the analytics, adjust inputs (i.e., query & explore the “why,” “how,” and “what-if”
   scenarios), and receive adaptive feedback.
5. To evaluate the effectiveness of the implemented InXLA system

Ideal Candidate

• Background in machine learning, data science, XAI or related fields.
• Experience or interest in working with educational data, particularly CBA.
• Programming skills (for data handling and prototyping).
• Background in Web Development / Dashboards -Valuable in building interactive
  explanation dashboards
• Enthusiastic about solving complex problems and contributing to impactful
  research