-
Gather details of other library projects that have auto-generated metadata from documents. Include the name of the organization, a summary of what they did, models and services used, etc. Anything t…
-
For users and especially developers getting familiar with the Skosmos codebase, it would help a lot to set up browsable API documentation generated using e.g. [phpDocumentor](https://www.phpdoc.org/).…
-
Is there any plan to support Python 3.10?
-
We could have a UI for administering most/all aspects of an Annif installation. It would most likely be a single-page dynamic web app that uses the REST API to perform admin functions.
-
A way of performing batch searches for indexed databases would be useful. A standard way of doing that, with tool support (OpenRefine at least), would be to provide a Reconciliation Service API: https…
-
Our workflow templates are using old action versions, we should update them to use the latest version instead:
https://github.com/step-security/secure-repo/blob/main/workflow-templates/scorecards.yml…
-
I would like to see the suggestions that are generated by Yake stored somewhere. Especially when using the webserver or the API. Currently showing the suggestions only works through the command line p…
-
The existing evaluation metrics (e.g. F1 and NDCG) are not ideal for evaluating hierarchical classification problems, e.g. assigning DDC or YKL classes, because they don't consider the hierarchy of th…
-
Just like with spaCy (see #374), we could add an analyzer that uses the [Stanza](https://stanfordnlp.github.io/stanza/) (formerly StanfordNLP) NLP toolkit for tokenization and especially lemmatization…
-
To speed up and ease the deposit into HAL and OSF, it would be good to prefill those fields with some reasonable suggestions and let authors decide what to keep.
http://annif.org/ by the National L…