Potential AI-based Use of Fuzzy Cognitive Mapping in Postgraduate Supervision in Higher Education

Abstract

Supervision in higher education is a complex and evolving process that necessitates adaptive, evidence-based decision-making to effectively guide postgraduate students. Conventional supervisory models often encounter difficulties in addressing uncertainties and the non-linear dynamics inherent in academic mentorship. This chapter examines the AI-based application of Fuzzy Cognitive Mapping (FCM) as an innovative framework to enhance supervisory practices by integrating expert insights, student progress data, and institutional guidelines within a structured yet flexible system. Employing a mental model approach, the study utilises fuzzy logic principles to simulate supervisory scenarios and assess causal relationships among critical factors, such as student motivation, research complexity, institutional support, and mentor–mentee engagement. The FCM-based framework enables supervisors to visualise interdependencies between variables, predict outcomes, and dynamically adjust mentoring strategies. Mixed methods, combining quantitative and qualitative data, are employed. Findings indicate that FCM enhances supervisory efficiency by promoting proactive interventions, improving communication, and supporting continuous monitoring of mentor–mentee relationships. Furthermore, the model advances a data-driven and transparent approach to supervision, minimising subjectivity while preserving contextual flexibility. By operationalising cognitive and computational intelligence, this chapter illustrates how FCM can bridge gaps between qualitative judgement and quantitative assessment in higher education supervision. The study contributes to emerging scholarship on artificial intelligence applications in academic contexts, underscoring the potential of cognitive modelling in improving student outcomes. It concludes by emphasising the necessity of empirical validation and the integration of adaptive mental models into institutional supervisory frameworks to strengthen postgraduate research management and mentoring effectiveness.