Using Large Language Models in Cluster Analysis in the Social Sciences
Bibliographic record
Abstract
The law, regulation, and policy of and for the digital economy can be viewed through different lenses. These include the formal approaches used by lawyers and academics through analysis by news businesses to content shared in video or audio form. Understanding the commonality and differences between the view through each of the lenses requires coordinated data sources. The International Digital Policy Observatory (IDPO) was created to develop a dataset across a variety of sources. This article demonstrates a novel methodological approach that uses data from the IDPO to analyze the interaction between different data sources. It does this using artificial intelligence regulation as an example and combines Gaussian Mixture Model (GMM) clustering techniques with Large Language Models (LLMs) for interpretable cluster naming to identify themes flowing from the data. It sets out the thematic outcomes in the context of each of the data source types to illustrate the method’s utility. This article’s primary contribution is methodological, presenting a scalable and interpretable workflow for analyzing large, multi-source text datasets in social science research. The clustering approach used is likely to be helpful in the analysis of text metadata in other large datasets.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.023 | 0.075 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.004 |
| Bibliometrics | 0.007 | 0.008 |
| Science and technology studies | 0.003 | 0.004 |
| Scholarly communication | 0.006 | 0.005 |
| Open science | 0.004 | 0.005 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".