in with
Tutorial

KorAP is a new Corpus Analysis Platform, suited for large, multiply annotated corpora and complex search queries independent of particular research questions.

New to KorAP? Please check out our guided tour or our tutorial!

News & Announcements

New KorAPClient version for Python published

To upgrade to v0.7.1 on Linux/macOS run python3 -m pip install -U KoraPClient and on Windows py -m pip install -U KoraPClient. The new version implements the changes of the corresponding R library and also includes a preliminary, still slow but powerful function for collocation analysis.

New RKorAPClient version published

To update to v0.7.1 in RStudio, go to Tools -> Check for Package Updates, select RKorAPClient and click Install Updates. The new version includes, among other things, a bug fix for searches with umlauts on Windows and a preliminary, still very slow but very powerful collocation analysis function (see also News).

KorAP has a new logo!

In the course of the relaunch of the IDS web pages KorAP also introduces a new logo - developed by Norbert Cußler-Volz. Just a first step - in the next couple of months the user interface will be modernized step by step and will adapt in parts to other IDS services.

New export feature for search results

Thanks to a newly integrated plugin, it is now possible to export search results directly via KorAP's graphical user interface. For this purpose, an Export icon appears above the hit list on the right after a successful search, which leads to a selection of initially three different export formats.

In addition, we have expanded the documentation in this version and, for example, introduced a short publication list as a citation aid for using KorAP.

dereko-korap-2021-I with over 21.3 billion tokens available via KorAP

In addition to the content of theeko-korap-2020-I, the virtual corpus contains new newspaper and journal volumes from 2021. See the detailed breakdown of the composition of dereko-korap-2021-I.

KorAP client library for Python released

DeReKo or KorAP can now be queried from the programming language Python. More information and installation instructions are available on PyPI or on GitHub.

dereko-korap-2020-I with over 20.6 billion tokens available via KorAP

The virtual corpus contains W-gesamt of COSMAS II (DeReKo-2020-I), Wikipedia corpora with articles and talk pages, and more corpora from DeReKo-2020-I. See the detailed breakdown of the composition of dereko-korap-2020-I.

KorAP is developed at the Leibniz Institute for the German Language, member of the Leibniz Association. The separated modules are being published as open source at GitHub.