Student Modeling — AI Ed Wiki

Student modeling — the broad practice of representing learner characteristics including knowledge, skills, affective states, engagement, and preferences in computational form. Student modeling is the foundation upon which adaptive and personalized AI in education systems are built.

Student modeling is the umbrella term encompassing all ways of representing learners in AI systems. Where Knowledge Tracing specifically tracks cognitive knowledge over time, student modeling includes affective modeling, engagement tracking, learning style inference, and preference learning. A complete student model answers: who is this learner, what do they know, how do they feel, and what do they need next?

Dimensions of student modeling

Cognitive modeling: Knowledge Tracing and Cognitive Diagnosis systems track what students know and their skill mastery

Affective modeling: Affective Computing and Affective Tutoring detect and respond to student emotions and motivation

Behavioral modeling: Learning Analytics systems infer engagement patterns from clickstream and interaction data

Proficiency modeling: Personalized Learning and Adaptive Learning platforms use student models to sequence content

LLM-era student modeling

Recent advances use LLMs for richer student modeling. The HiLLM-CD framework represents students as proficiency trees, while multimodal approaches construct evidence-grounded knowledge representations from diverse data sources. LLMs enable automated model construction from educational text, reducing reliance on expert annotation.

Connections to other concepts

Student modeling feeds into Intelligent Tutoring (for problem selection), Personalized Learning (for content sequencing), and Learning Analytics (for dashboards and interventions). It connects fundamentally to Knowledge Tracing for cognitive tracking and to Adaptive Learning for system adaptation.