Pattern-based knowledge component extraction from program ASTs

Research Method · All research methods
Evidence · 1 claim (1 for) · 1 study (1 design), q2 · 0 of 1 report an effect size · 1 claim rests on one study

Description

The framework defines knowledge components as "recurring, semantically meaningful structural patterns based on important subtrees from program Abstract Syntax Trees (ASTs)". Discovery proceeds in four steps: SANN-guided subtree extraction trained on topic labels, normalization that anonymizes identifiers and canonicalizes syntactic alternatives, context-aware encoding of normalized subtree sequences with a β-VAE, and K-means clustering of latent vectors into KCs. Activities are then mapped to IDF-weighted knowledge vectors and compared by cosine similarity, yielding finer-grained, implementation-level signals such as guarded loop traversals and accumulator updates.

Accounts

• Pattern-based knowledge component framework for representing programming learning activities: The framework defines knowledge components as "recurring, semantically meaningful structural patterns based on important subtrees from program Abstract Syntax Trees (ASTs)". Discovery proceeds in four steps: SANN-guided subtree extraction trained on topic labels, normalization that anonymizes identifiers and canonicalizes syntactic alternatives, context-aware encoding of normalized subtree sequences with a β-VAE, and K-means clustering of latent vectors into KCs. Activities are then mapped to IDF-weighted knowledge vectors and compared by cosine similarity, yielding finer-grained, implementation-level signals such as guarded loop traversals and accumulator updates. (Muntasir Hoq et al. (2026))

Claims

• Pattern-based KC similarity retrieves bundle-aligned programming learning items better than code-embedding and other KC baselines in both retrieval directions [+M]

Related Research Methods

• Tool-supported workflow for identifying, clustering, and visualizing common modelling mistakes as bug fixing transformations

Key Sources

• Muntasir Hoq, Griffin Pitts, Zhangqi Duan, Arun Balajiee Lekshmi Narayanan, Mohammad Hassany, Andrew Lan, Peter Brusilovsky and Bita Akram. (2026). Automated Recommendation of Programming Learning Content Using Pattern-based Knowledge Components. CSEDM'26: 10th Educational Data Mining in Computer Science Education Workshop. https://educationaldatamining.org

Page metadata

Field	Value
type	research-method
id	pattern-based-knowledge-component-extraction-from-program-asts
title	Pattern-based knowledge component extraction from program ASTs
description	The framework defines knowledge components as "recurring, semantically meaningful structural patterns based on important subtrees from program Abstract Syntax Trees (ASTs)".
status	review
generated	by process:settle-candidates on 2026-10-11
trust	unverified — no one has confirmed this page's content

Frontmatter is the page's machine-readable record — see the Schema & Guide.