Solid arrows = proposed prerequisite edges (thickness = prior); dotted = is-a. Node size = number of book units tagged. Drag to move, scroll to zoom.
A KC that an exercise tests or the text assumes before any unit teaches it. never taught in the book · taught later.
Prior = mean of textbook order, counterfactual judgment, and zero-shot scores (0–1). The last column is the book's own order: does the prerequisite appear first?
| from | to | prior | order |
|---|
Number of distinct KCs an exercise requires (Mihaylova 2026's KC-load metric).
| unit | kind | section | KCs |
|---|
Hand-written prerequisite edges in the course knowledge base, and whether the pipeline mapped and re-proposed them.
| from | to | mapped | proposed |
|---|
Method: ontology-guided tagging (Micheli 2025 Python ontology as the seed for chapters 0–2; LLM-proposed typed hierarchy for chapters 3–4), one annotation pass per unit by Claude, score ≥ 4 counts as present; roles teaches / assumes / exercises; leaps and KC load follow Smith 2026 and Mihaylova 2026; edges carry a prior, not a verdict. No student data yet; learning-curve validation is Phase 2.