Track the frontier, from condensed matter to cosmology.
Search papers, labs, and topics across Lattice.
This paper addresses the challenge of aligning topic models across different media by introducing a reproducible framework that situates various corpora within a unified topic space based on a predefined taxonomy. Utilizing guided BERTopic, the authors score and refine topics against the IPTC Media Topics taxonomy, ultimately consolidating them into a coherent structure of seventeen parent topics. The results show that this method significantly enhances topic coverage and consistency compared to traditional zero-shot approaches, demonstrating its effectiveness in cross-source topic modeling.
A novel framework for cross-source topic modeling reveals that guided topic alignment can dramatically improve coverage and consistency across diverse media corpora.
Comparing topic attention across different media is hindered by a fundamental modelling problem: topic models fitted separately to each corpus produce corpus-specific topic spaces that cannot be aligned directly. This paper presents a reproducible framework that places corpora in a single shared topic space defined by a taxonomy. Discovered topics are obtained with guided BERTopic, scored against the ninety-four IPTC Media Topics' taxonomy topics (level-1) through weighted keyword and target centroids, and then collapsed upward to seventeen IPTC parent topics by a maximum-similarity rule. The framework was developed and selected on a controlled New York Times 2011 corpus through a narrowing sequence: a broad model screen, a focused mapping refinement, a strict finalist comparison, a target-construction ablation, and a threshold calibration. In this corpus, the guided family retained substantially stronger mapped coverage than a zero-shot benchmark under stricter assignment thresholds, a parent-enriched target construction improved both coverage and parent consistency, and coverage declined gradually rather than collapsing as the assignment threshold was tightened. The contribution is an externally anchored method for constructing a shared topic space that enables reproducible cross-source topic comparison.