The DOI matches its Crossref record.
Kratt: Developing an Automatic Subject Indexing Tool for the National Library of Estonia
Open accessAbstract
Manual subject indexing in libraries is a time-consuming and costly process and the quality of the assigned subjects is affected by the cataloger’s knowledge on the specific topics contained in the book. Trying to solve these issues, we exploited the opportunities arising from artificial intelligence to develop Kratt: a prototype of an automatic subject indexing tool. Kratt is able to subject index a book independent of its extent and genre with a set of keywords present in the Estonian Subject Thesaurus. It takes Kratt approximately one minute to subject index a book, outperforming humans 10–15 times. Although the resulting keywords were not considered satisfactory by the catalogers, the ratings of a small sample of regular library users showed more promise. We also argue that the results can be enhanced by including a bigger corpus for training the model and applying more careful preprocessing techniques.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“However, professional catalogers rated the quality of the assigned subjects as unsatisfactory, citing many inaccurate or missing terms-although a small sample of regular library users found the results somewhat more useful for discovery.”
From A Skill-Based AI Agentic Pipeline for Library of Congress Subject Indexing (2026) · cites this paper as Asula et al., 2021
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Unlike the XMTC approaches of Asula et al. (2021) and Suominen et al. (2025), the system does not require training data or corpus-level pattern learning.”
From A Skill-Based AI Agentic Pipeline for Library of Congress Subject Indexing (2026) · cites this paper as Asula et al. (2021)
The cited source says what the claim attributes to it.
In the citing paper
“The system's training data suffered from severe label sparsity: the median frequency of unique labels was only 2 across the training set, meaning most subject terms had too few examples for robust learning.”
From A Skill-Based AI Agentic Pipeline for Library of Congress Subject Indexing (2026) · cites this paper as Asula et al., 2021
The cited source says what the claim attributes to it.
In the citing paper
“Kratt processed books approximately 10-15 times faster than human catalogers, completing subject indexing in about one minute per title.”
From A Skill-Based AI Agentic Pipeline for Library of Congress Subject Indexing (2026) · cites this paper as Asula et al., 2021
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cited by
Sorted by citations
Research on Large Language Model-Based Bibliographic Cataloging Agent in the CNMARC Context
Tan, Z., Yang, X., Chen, Q., & Chen, T. (2026) · Publications · doi:10.3390/publications14010019 · Cited by 2
“In addition, the National Library of Estonia, in collaboration with Texta OÜ, developed Kratt, an automatic subject indexing tool that utilizes local corpora and machine learning techniques to enable rapid subject assignment for documents (Asula, 2021).”
A Skill-Based AI Agentic Pipeline for Library of Congress Subject Indexing
Chow, E. H. C. (2026) · arXiv · doi:10.48550/arxiv.2605.03537 · Cited by 1
“The National Library of Estonia developed Kratt, a prototype automatic subject indexing tool that used page-level logistic regression classifiers to assign terms from the Estonian Subject Thesaurus (Asula et al., 2021).”
Metadata
- Authors
- Asula, M., Makke, J., Freienthal, L., Kuulmets, H.-A., & Sirel, R.
- Affiliations
- Texta OÜ, Tallinn, Estonia
- National Library of Estonia (EE)
- Corresponding author
- Asula, Marit
- ORCID
- Kuulmets, Hele-Andra
- Venue
- Cataloging & Classification Quarterly
- Publisher
- Informa UK Limited
- Published
- 2021-11-01
- Volume
- 59
- Issue
- 8
- Pages
- 775-793
- ISSN
- 0163-9374 (ISSN-L)
- 1544-4554
- OpenAlex
- W3208114150
- Cited by
- 14
- References
- 0
- Citations by year
- 2026: 5 · 2025: 4 · 2024: 3 · 2023: 2
- FWCI
- 1.13
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- Cataloging & Classification Quarterly (publishedVersion, public-domain)
- arXiv (Cornell University) (submittedVersion)
- Zenodo (CERN European Organization for Nuclear Research) (submittedVersion, cc-by)
- arXiv (Cornell University)
- Funders
- European Regional Development Fund
- Ministry of Education and Research (EKTR3)
- European Union’s Horizon 2020 research and innovation programme (825153)
- Topics
- Physical Sciences › Computer Science › Information Systems › Library Science and Information Systems
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Keywords
- automatic subject indexing, library cataloging, training corpus size, text preprocessing
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
