Knowledge Graph

Knowledge graph — a structured representation of concepts and their relationships used to model domain knowledge, student understanding, and learning dependencies in AI in education systems. Knowledge graphs enable AI systems to reason about what students know, what they need to learn next, and how concepts relate to each other.

Questions to Consider

• A knowledge graph captures not just concepts but their relationships — prerequisites, similarity, hierarchy. Why might knowing how concepts relate be more useful to an adaptive system than a flat list of skills?
• How are prerequisite relationships in a domain like yours? Can you think of a topic where students routinely struggle because they're missing a foundational concept the graph would reveal?
• The page describes using knowledge graphs to detect knowledge gaps — where learners are missing foundational concepts. How might surfacing that gap change what an AI tutor decides to teach next?
• Knowledge graphs can be built manually or automatically by LLMs from educational text. What are the risks of letting an AI construct the concept structure that a tutor will then reason over?
• If knowledge graphs provide the domain structure that AI agents reason over, what happens to trust and accuracy when the graph itself contains an error or a biased relationship?
• A knowledge graph is described as the structural backbone enabling fine-grained diagnosis and personalized paths. In your own teaching or design, what would you need a knowledge graph of your subject to capture — and what would it leave out?

Introduction

Knowledge graphs provide the structural backbone for many intelligent education systems. Unlike flat lists of skills or concepts, knowledge graphs capture prerequisite relationships, similarity, and hierarchical organization — essential for Adaptive Learning, Knowledge Tracing, and Learner Modeling and Adaptive Instruction.

How knowledge graphs are used in AIED

Knowledge graphs are a recurring structural mechanism across the knowledge base's AIED research, serving several distinct roles:

• Knowledge Tracing models use concept graphs to propagate student proficiency estimates across related skills, improving prediction accuracy when data is sparse.
• Learner Modeling and Adaptive Instruction systems leverage knowledge graphs to represent what learners know in a semantically meaningful way, enabling fine-grained diagnosis.
• Adaptive Learning platforms use prerequisite graphs to sequence content and recommend personalized learning paths.
• Algorithm choice over the graph changes outcomes. In G4L, propagating mastery through an Evolving Knowledge Space Graph with Bayesian knowledge propagation produced +24% measured knowledge (0.717 → 0.887), against +5% for Knowledge Space Theory and +1% for Weighted Distance Dependent Induction (Csépányi-Fürjes & Kovács, 2026).
• Cognitive Diagnosis frameworks like HiLLM-CD construct concept trees from educational text using LLMs, eliminating manual annotation.
• Knowledge-graph-augmented tutoring: ITAS uses a knowledge graph of quantum concepts (with explicit prerequisite relationships) to drive a multi-agent tutoring system, traversing the graph to select next topics for counterintuitive material.
• Curriculum and course modeling: CourseGraph compares CS course structures across institutions using graph representations; Learnity graphs model lifelong learning pathways.
• Prerequisite-relation learning: ProPrL learns prerequisite relations among concepts, formalizing the edges that knowledge graphs encode.
• Learnable prerequisite discovery replaces expert priors. A survey synthesis by Xu et al. (2026) finds that graph-based tracers can learn prerequisite structure from response data during training, replacing expert-crafted priors. GKT parameterizes a learnable adjacency matrix end-to-end, SKT-L learns prerequisites while training, PKT applies L1 regularization to recover sparse prerequisite graphs evaluated against ground-truth structures with F1 scores, and PSI-KT's variational Bayesian framework infers directed prerequisite graphs that align with human-annotated prerequisites better than baseline models.
• Knowledge-gap detection: Knowledge gap detection uses graph-based reasoning in AI teaching assistants to identify where learners are missing foundational concepts.
• Multimodal AI and explainable reasoning: multimodal knowledge graphs extend graph structure across content modalities; fair and explainable recommendations combine knowledge-graph embeddings with sequential modeling (a hybrid HKG-GRU framework).
• Instructionally structured graphs for resource recommendation: Liu, Sun & Song (2026) decompose each teaching-resource entity into four instructional dimensions (teaching context, cognitive level, technological feature, cultural adaptability), compute user-dependent semantic similarity over those dimensions, and fuse it with collaborative filtering via an ability- and progress-aware coefficient — encoding pedagogical structure directly into the recommendation signal rather than treating resources as consumption items.
• Ontology-based knowledge bases: Ivanova (2026) proposes a layered, hybrid knowledge-base architecture grounded in description logic that replaces the classic ITS single-ontology models with systems of mapped ontologies — adding procedural (rule-based), probabilistic/fuzzy, and ML-extracted implicit knowledge — plus a metadata framework for describing, discovering, and reusing educational ontologies.
• Scaffolding and writing: Veriforge and visual query tracing apply graph-based structure to narrative drafting and declarative-logic learning.
• Human-curated literary graphs, and what an audit exposes: Incipit graphs literary premises — 1,455 axiom records, 1,464 mappings to 149 works, and 472 typed relationships — with language models proposing candidate formulations that human curators selected and grounded. Its recomputed audit is as instructive as its structure: every endpoint resolves and no duplicate or self-link remains, yet 1,448 of the 1,455 axioms map to exactly one work (so cross-work reuse is sparse), the context taxonomy cannot separate its two context types, and no provenance record survives, leaving the snapshot unable to reconstruct its own pipeline. Structural validity is not interpretive quality, and a curated graph without provenance can be neither audited nor refreshed.

LLM-driven knowledge graph construction

Recent research explores using LLMs to automatically construct knowledge graphs from educational content. The HiLLM-CD framework uses multi-agent LLM pipelines to generate exercise-concept links and hierarchical concept trees, reducing reliance on expert annotation. This connects to broader Generative AI applications in curriculum design and automated content organization, and to RAG (Retrieval-Augmented Generation) (retrieval-augmented generation), where graph-structured knowledge can improve retrieval quality over flat similarity search.

Relationship to other concepts

Knowledge graphs connect to Learning Design (defining what to teach), Curriculum Design (how to sequence it), and Learning Analytics (extracting insights from student interaction data). They are foundational to Intelligent Tutoring systems that need structured representations of educational domains. As AI agents become more common in education, knowledge graphs provide the domain structure that agentic systems reason over — a pattern seen in ITAS and knowledge-gap-detection teaching assistants.

Common Misconceptions

"A knowledge graph is a concept map with better tooling." The distinguishing claim is relational and computational: a graph is useful because prerequisites, similarity and hierarchy are explicit enough for an algorithm to traverse, not because it draws connected nodes. What follows from that is the operational question the page keeps returning to — which traversal or propagation algorithm runs over the graph — because that choice is where the instructional effect actually lives.

"Having the graph is what improves outcomes." The algorithm over the graph changes the outcome. In G4L, propagating mastery through an Evolving Knowledge Space Graph with Bayesian knowledge propagation produced +24% measured knowledge (0.717 → 0.887), against +5% for a Knowledge Space based comparison. Same general idea of a structured knowledge space, materially different result depending on the propagation method. Buying or building a graph is the beginning of the design problem, not its solution.

"An LLM-built graph that passes a structural audit is a validated graph." Structural validity and instructional validity come apart in exactly this way. The recomputed audit of one such graph found that every endpoint resolves and no duplicate or self-link remains — a clean structure — yet 1,448 of its 1,455 axioms map to exactly one work, so cross-work reuse is sparse, its context taxonomy cannot separate its two context types, and central claims remain unproven. Dead links and duplicates are the failures an automated audit catches; thin reuse and undifferentiated categories are the ones it waves through.

Implications for Practice

For instructors

• Ask what the graph is used for before asking whether it is accurate: recommendation, prerequisite sequencing, or diagnosis. The evidence differs sharply by use, and prerequisite sequencing carries the strongest result.
• When a tool claims adaptive sequencing, ask for the measured comparison against a non-adaptive baseline rather than a description of the knowledge structure.
• Expect sparsity in cross-work reuse, and treat a graph whose entities map one-to-one onto single sources as a catalogue with edges rather than as an integrated map.

For developers

• Treat the propagation algorithm as a first-class design decision and report the algorithm as part of the intervention; the +24% versus +5% gap is attributable to it.
• Audit beyond resolvability. Test whether categories are separable and whether entities are reused across works, because structural cleanliness consistently coexists with instructional thinness.
• Validate the graph against learning outcomes, not against graph statistics.

For researchers

• Report graph-baseline contrasts at the algorithm level. "Knowledge-graph-based" is not a treatment; the traversal strategy is.
• Report axiom-to-source mappings and taxonomy separability alongside precision and recall, since the audit failures here were invisible to conventional graph metrics.

Connected Concepts

• Adaptive Learning
• Knowledge Tracing
• Intelligent Tutoring
• Cognitive Diagnosis
• Learner Modeling and Adaptive Instruction
• Learning Analytics
• Curriculum Design
• Learning Design
• Generative AI
• Large Language Models (LLMs)
• RAG (Retrieval-Augmented Generation)
• Agentic AI
• Technologies — Umbrella: AI technologies and techniques (models, LLM training, robotics, RAG, agentic)
• Recommender Systems and Learning Paths

Connected Articles

• Literary Axioms: A Conceptual Framework for Literary Creation, Interpretation, and Evaluation — A curator-built graph of 1,455 literary axioms with a structural audit and no provenance record (Liu & Zhao 2026)
• Ontology-based layered hybrid AI-driven knowledge model for personalized e-learning — Ontology-based layered hybrid knowledge model for personalized e-learning
• Rethinking Higher Education: From Fixed Curricula to Learnity Graphs — Learnity graphs for lifelong learning
• VeriForge: Mitigating Latent Knowledge Gaps in Narrative Drafting via Mixed-Initiative Scaffolding — Veriforge: narrative-drafting scaffolds
• From Prototype to Classroom: An Intelligent Tutoring System for Quantum Education — Quantum education intelligent tutoring (ITAS)
• Evidence-Grounded Multimodal Knowledge Graph Construction for Multi-Lecture Educational Reasoning — Multimodal knowledge graphs for educational reasoning
• CourseGraph: Finding overlaps and differences in Computer Science courses across universities — CourseGraph: CS course comparison
• ProPRL: Property-Aware Prerequisite Relation Learning in Educational Knowledge Graphs — ProPrL: prerequisite-relation learning
• Detecting Knowledge Gaps from Conversational AI Interactions Using Curriculum Prerequisite Graphs — Knowledge-gap detection in AI teaching assistants
• Evaluating a Visual Query Tracer and Builder for Learning Declarative Logic Programming — Visual query tracer for declarative logic learning
• Fair and explainable educational recommendations with a hybrid Graph-GRU framework — Fair and explainable educational recommendations
• A Hybrid Collaborative Filtering and Knowledge Graph-Based Cross-Domain Recommendation Method for Multimodal Teaching Resources — Hybrid CF–KG cross-domain recommendation for multimodal teaching resources
• HiLLM-CD: LLM-Enhanced Hierarchical Cognitive Diagnosis — HiLLM-CD: LLM-driven cognitive diagnosis
• Intelligent tutoring in dynamic domains: a graph-based system for comparative analysis of adaptive algorithms with intuitionistic fuzzy logic and forgetting — Graph-Based Intelligent Tutoring for Dynamic Domains (2026)
• CogEvol: Towards Efficient and Reliable Learning Environment Generation — CogEvol: Learning Environment Generation