We are pleased to announce the release of INCEpTION 0.9.0.
INCEpTION is an extensible web-based collaborative environment
for text annotation. It integrates machine-learning-based annotation
support, knowledge base management and corpus building into
a single comprehensive platform.
This is a feature release.
Notable changes since INCEpTION 0.8.4
- Added display of images from image features in the comment popup
- Added display of label/description from KB features in the comment popup
- Added learning simulation for recommenders to recommender settings (project settings)
- Added learning curve plot to the recommender sidebar (annotation page)
- Added button to allow annotators to leave a project
- Added support for FTS in Fuseki-based remote KBs
- Added support for entity search on the official Wikidata SPARQL endpoint
- Added ability to configure max. results for ElasticSearch document repos
- Added ability to configure the default text field for ElasticSearch document repos
- Added labels to concepts in the example wine KB to make it searchable
- General performance improvements for text annotation
- General performance improvements for knowledge base access
- Improved entity (concept) selection auto-complete field
- Improved loading times on the knowledge base page
- Improved quality score calculation for recommenders
- Fixed some problems accessing Wikidata in reified mode
- Fixed error message spamming the logs when indexing large annotations
- ... and many more improvements and bug fixes under the hood
If you discover problems with this release,
please report them to us via our issue tracker [2].
Thanks to the developer team go to: @reckart, @UWinch, @Sibgha360, @jcklie, @MdeBoer95, @az79nefy
Thanks for contributions go to: @chaitraliagashe, @wpmp
Also thanks to all issue reporters, beta testers, feature suggesters, and users!
Download
The new version can be downloaded from our website [3].
Important upgrade notes
An in-place upgrade from INCEpTION 0.8.4 to 0.9.0 is possible.
Back up your data before the upgrade [4].
Publication
An overview of INCEpTION can be found in
Klie, J.-C., Bugert, M., Boullosa, B., Eckart de Castilho, R. and Gurevych, I. (2018):
The INCEpTION Platform: Machine-Assisted and Knowledge-Oriented Interactive Annotation.
In Proceedings of System Demonstrations of the 27th International Conference on Computational Linguistics (COLING 2018), Santa Fe, New Mexico, USA
PDF and BibTeX as well as more publications on specific features are
available from our website [5].
License and System requirements
INCEpTION is provided as open source software under the Apache License version 2.
Running INCEpTION requires Java version 8 or higher.
By default, INCEpTION uses an embedded database which is sufficient for testing. However, the use of a MySQL database is recommended.
INCEpTION works best with Chrome and Safari.
-- The INCEpTION developer team
[1] https://github.com/inception-project/inception/issues?q=milestone%3A0.9.0
[2] https://github.com/inception-project/inception/issues
[3] https://inception-project.github.io/downloads/
[4] https://inception-project.github.io/releases/0.9.0/docs/admin-guide.html#_backup_your_data
[5] https://inception-project.github.io/publications/