Self‐organizing maps for latent semantic analysis of free‐form text in support of public policy analysis
Bibliographic record
Abstract
The huge amount of free‐form unstructured text in the blogosphere, its increasing rate of production, and its shrinking window of relevance, present serious challenges to the public policy analyst who seeks to take public opinion into account. Most of the tools which address this problem useXMLtagging and other Web 3.0 approaches, which do not address the actual content of blog posts and the associated commentary. We give a tutorial review of latent semantic analysis and the self‐organizing maps, as considered in this context, and show how to apply the self‐organizing map over a probabilistic latent semantic space to the problem of completely unsupervised clustering of unstructured text in such a way as to be entirely independent of spelling, grammar, and even source language. This provides an algorithm suitable for clustering free‐form commentary with a well‐structured test environment. The algorithm is applied to academic paper abstracts instead, treated as unstructured text as though they were blog posts, because this set of documents has a known ground truth. The algorithm constructs a word category map and a document map in which words with similar meaning and documents with similar content are clustered together.WIREs Data Mining Knowl Discov2014, 4:71–86. doi: 10.1002/widm.1112 This article is categorized under: Algorithmic Development > Web Mining Application Areas > Government and Public Sector Technologies > Structure Discovery and Clustering
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 | 0.003 | 0.016 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.004 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.002 |
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".