Relations sémantiques pour l'indexation automatique. Définition d'objectifs pour la détection automatique
Bibliographic record
Abstract
L'accès aux documents numériques volumineux ou complexes peut être facilité par un index du style que l'on retrouve à la fin d'un livre, présentant schématiquement les concepts abordés dans le document et les liens que l'auteur a établi entre eux.Il peut s'avérer un outil précieux dans la fouille de documents.Le travail de recherche décrit ici vise à identifier les relations sémantiques présentes dans les index de livre produits manuellement pour déterminer lesquelles peuvent être dérivées automatiquement.Pour ce faire, sept index ont été examinés.Les observations relevées permettent de distinguer deux types de relations : celles pour lesquelles l'analyse du document en main fournit suffisamment d'informations, et celles pour lesquelles des ressources terminologiques externes sont nécessaires.Des pistes pour le développement d'un système d'indexation automatique de monographies sont ainsi identifiées.ABSTRACT.Access to large or complex digital documents can be facilitated by a so-called « back-of-the-book index », which presents schematically the concepts discussed in the document and links made between them by the author.It can thus be a very useful tool to explore document content.The research project described here pertains to an analysis of semantic relations expressed in manually-compiled indexes, in order to determine which could be derived automatically.Seven indexes were examined.The resulting observations suggest two types of relations: those that can be calculated simply from the document's content, and those for which external terminological resources are necessary.This has identified areas for further research into automatic back-of-the-book indexing.
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 | 0.010 | 0.028 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.003 |
| Bibliometrics | 0.009 | 0.008 |
| Science and technology studies | 0.003 | 0.005 |
| Scholarly communication | 0.013 | 0.016 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.010 | 0.008 |
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".