Kratt: Developing an Automatic Subject Indexing Tool for The National Library of Estonia

TL;DR

Kratt is an AI-based tool that automates subject indexing of books in Estonian libraries, significantly reducing time and showing potential for improved accuracy with further development.

Contribution

This paper introduces Kratt, a novel automatic subject indexing system for Estonian books, leveraging AI to improve efficiency and accuracy over manual cataloging.

Findings

Kratt indexes a book in about 1 minute, outperforming humans by 10-15 times.

User ratings suggest potential for improved keyword quality with more training data.

The system's performance can be enhanced with larger datasets and better preprocessing.

Abstract

Manual subject indexing in libraries is a time-consuming and costly process and the quality of the assigned subjects is affected by the cataloguer'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 1 minute to subject index a book, outperforming humans 10-15 times. Although the resulting keywords were not considered satisfactory by the cataloguers, 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.