Lingolol

0 posts0 participants0 posts today

Osma SuominenIn the first <a href="https://sigmoid.social/tags/LLMs4Subjects" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLMs4Subjects</a> challenge at the SemEval-2025 workshop, our <a href="https://sigmoid.social/tags/Annif" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Annif</a> team did very well!The challenge was to generate good quality subject indexing for bibliographic records in German & English using LLMs. We used LLMs for data preprocessing (translation & synthetic data) and Annif as the main suggestion engine. We ranked 1st and 2nd in quantitative and 4th in qualitative evaluations out of 14 teams!More info & preprints: <a href="https://groups.google.com/g/annif-users/c/b8kVy6XSzB4/m/JE6xBzSuEgAJ" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://groups.google.com/g/annif-users/c/b8kVy6XSzB4/m/JE6xBzSuEgAJ</a><a href="https://sigmoid.social/tags/subjectIndexing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#subjectIndexing</a> <a href="https://sigmoid.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a> <a href="https://sigmoid.social/tags/LLM" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LLM</a> <a href="https://sigmoid.social/tags/SemEval" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#SemEval</a>

Osma SuominenLast November we organized a survey for Annif users, and now the results have been published in the Doria repository of the National Library of Finland:<a href="https://urn.fi/URN:ISBN:978-952-84-1301-1" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://urn.fi/URN:ISBN:978-952-84-1301-1</a>The report includes an overview of:- The vocabularies and datasets that are used with Annif - The workflows that Annif is integrated with - The problems Annif users are facing<a href="https://sigmoid.social/tags/Annif" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Annif</a> <a href="https://sigmoid.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a> <a href="https://sigmoid.social/tags/opensource" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#opensource</a> <a href="https://sigmoid.social/tags/SKOS" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#SKOS</a> <a href="https://sigmoid.social/tags/libraries" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#libraries</a> <a href="https://sigmoid.social/tags/subjectIndexing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#subjectIndexing</a>

Osma SuominenVersion 1.3 of the automated subject subject indexing tool <a href="https://sigmoid.social/tags/Annif" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Annif</a> has been released!This release introduces support for the EstNLTK analyzer for better Estonian lemmatization 🇪🇪, optimizations to the MLLM backend, as well as maintenance and bug fixes, including better file permissions in multi-user environments.<a href="https://github.com/NatLibFi/Annif/releases/tag/v1.3.0" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://github.com/NatLibFi/Annif/releases/tag/v1.3.0</a><a href="https://sigmoid.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a> <a href="https://sigmoid.social/tags/machinelearning" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#machinelearning</a> <a href="https://sigmoid.social/tags/opensource" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#opensource</a> <a href="https://sigmoid.social/tags/code4lib" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#code4lib</a> <a href="https://sigmoid.social/tags/libraries" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#libraries</a> <a href="https://sigmoid.social/tags/subjectindexing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#subjectindexing</a> <a href="https://sigmoid.social/tags/SKOS" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#SKOS</a> <a href="https://sigmoid.social/tags/classification" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#classification</a> <a href="https://sigmoid.social/tags/Estonian" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Estonian</a> <a href="https://sigmoid.social/tags/eesti" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#eesti</a>

Osma SuominenThe automated subject indexing tool <a href="https://sigmoid.social/tags/Annif" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Annif</a> version 1.2 has been released! This release adds API and CLI functionality for language detection (based on the <a href="https://sigmoid.social/tags/Simplemma" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Simplemma</a> library), automated download of NLTK data packages, full support for Python 3.12, a few bug fixes and upgraded dependencies.<a href="https://sigmoid.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a> <a href="https://sigmoid.social/tags/machinelearning" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#machinelearning</a> <a href="https://sigmoid.social/tags/opensource" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#opensource</a> <a href="https://sigmoid.social/tags/code4lib" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#code4lib</a> <a href="https://sigmoid.social/tags/libraries" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#libraries</a> <a href="https://sigmoid.social/tags/subjectindexing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#subjectindexing</a> <a href="https://sigmoid.social/tags/SKOS" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#SKOS</a> <a href="https://sigmoid.social/tags/classification" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#classification</a> <a href="https://github.com/NatLibFi/Annif/releases/tag/v1.2.0" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://github.com/NatLibFi/Annif/releases/tag/v1.2.0</a>

v_i_o_l_a"An Experiment with the Use of ChatGPT for LCSH Subject Assignment on Electronic Theses and Dissertations" <a href="https://doi.org/10.1080/01639374.2024.2394516" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://doi.org/10.1080/01639374.2024.2394516</a> <a href="https://openbiblio.social/tags/ChatGPT" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ChatGPT</a> <a href="https://openbiblio.social/tags/subjectindexing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#subjectindexing</a>

Sophie SchneiderHier die Folien zu meinem Vortrag “KI-gestützte Inhaltserschließung im Projekt Mensch.Maschine.Kultur” heute auf der <a href="https://fedihum.org/tags/GBV" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#GBV</a> Verbundkonferenz: <a href="https://zenodo.org/doi/10.5281/zenodo.13365610" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://zenodo.org/doi/10.5281/zenodo.13365610</a><a href="https://fedihum.org/tags/automaticindexing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#automaticindexing</a> <a href="https://fedihum.org/tags/subjectindexing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#subjectindexing</a> <a href="https://fedihum.org/tags/machinelearning" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#machinelearning</a> <a href="https://fedihum.org/tags/LIS" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LIS</a>

Sophie Schneiderthe (hi)story goes on... We just published machine-readable vocabulary files for the historical classification system "Alter Realkatalog" (ARK) <a href="https://openbiblio.social/@stabi_berlin" class="u-url mention" rel="nofollow noopener noreferrer" target="_blank">@stabi_berlin</a> as well as corresponding <a href="https://fedihum.org/tags/annif" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#annif</a> (automatic subject indexing) models, trained on our ARK vocabulary and metadata: <a href="https://zenodo.org/doi/10.5281/zenodo.13301019" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://zenodo.org/doi/10.5281/zenodo.13301019</a><a href="https://huggingface.co/collections/SBB/ark-annif-models-66bc8c4dea476bea05a6b491" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://huggingface.co/collections/SBB/ark-annif-models-66bc8c4dea476bea05a6b491</a><a href="https://fedihum.org/tags/automaticindexing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#automaticindexing</a> <a href="https://fedihum.org/tags/subjectindexing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#subjectindexing</a> <a href="https://fedihum.org/tags/machinelearning" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#machinelearning</a> <a href="https://fedihum.org/tags/LIS" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#LIS</a> From: <a href="https://mastodon.social/@jrglmn" class="u-url mention" rel="nofollow noopener noreferrer" target="_blank">@jrglmn</a> <a href="https://mastodon.social/@jrglmn/112830851972092378" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://mastodon.social/@jrglmn/112830851972092378</a>

v_i_o_l_a"Automating subject indexing at ZBW: making research results stick in practice" by <a href="https://openbiblio.social/@SemAntiKast" class="u-url mention" rel="nofollow noopener noreferrer" target="_blank">@SemAntiKast</a>: <a href="https://doi.org/10.53377/lq.13579" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://doi.org/10.53377/lq.13579</a> <a href="https://openbiblio.social/tags/subjectindexing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#subjectindexing</a> <a href="https://openbiblio.social/tags/ZBW" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ZBW</a>

Sophie SchneiderIn an internal workshop <a href="https://openbiblio.social/@stabi_berlin" class="u-url mention" rel="nofollow noopener noreferrer" target="_blank">@stabi_berlin</a> we asked our colleagues working on subject indexing: How can we employ automation to effectively support subject indexing? Read about the responses we received: <a href="https://mmk.sbb.berlin/2023/09/05/workshop-semi-automatic-subject-indexing-at-the-berlin-state-library/?lang=en" rel="nofollow noopener noreferrer" translate="no" target="_blank">https://mmk.sbb.berlin/2023/09/05/workshop-semi-automatic-subject-indexing-at-the-berlin-state-library/?lang=en</a> <a href="https://mastodon.social/tags/automaticindexing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#automaticindexing</a> <a href="https://mastodon.social/tags/subjectindexing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#subjectindexing</a> <a href="https://mastodon.social/tags/libraries" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#libraries</a> <a href="https://mastodon.social/tags/ai" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#ai</a>

Osma SuominenThe automated subject indexing tool <a href="https://sigmoid.social/tags/Annif" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#Annif</a> has reached version 1.0 after six years of development!The 1.0 release signals that it is ready for production use, and that from now on, even more emphasis will be placed on backwards compatible releases to ease future upgrades. Also, there is improved CLI support, preliminary support for Python 3.11, lots of bug fixes and small improvements.<a href="https://github.com/NatLibFi/Annif/releases/tag/v1.0.0" rel="nofollow noopener noreferrer" target="_blank">https://github.com/NatLibFi/Annif/releases/tag/v1.0.0</a><a href="https://sigmoid.social/tags/libraries" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#libraries</a> <a href="https://sigmoid.social/tags/machinelearning" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#machinelearning</a> <a href="https://sigmoid.social/tags/classification" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#classification</a> <a href="https://sigmoid.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#AI</a> <a href="https://sigmoid.social/tags/subjectindexing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#subjectindexing</a> <a href="https://sigmoid.social/tags/SKOS" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#SKOS</a> <a href="https://sigmoid.social/tags/code4lib" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#code4lib</a>

Recent searches

Search options

Administered by:

Server stats:

#subjectindexing