Tool made to add main subject statements to items based on the title using a home-brewed CirrusSearch-based Named Entity Recognition algorithm. bild The tool running in PAWS adding manually found main subject QIDs


This tool has the following features

  • adding a list of main subjects to items (for now only scholarly articles are supported)
  • automatically extracting n-grams from labels of 10.000 articles (this is not that powerful because users know better than scikit what subjects are meaningful to have on our scientific articles)


During the development of this tool the author got a help multiple times from Jan Ainali and Jon Søby with figuring out how to query the API using the CirrusSearch extensions and to remove more specific main subjects from the query results.

A special thanks also to Magnus Sälgö for his valuable input and ideas, e.g. to search for aliases also.


Clone the repo and run

pip install -r requirements.txt

to install all requirements.


The tool runs in PAWS with no known issues.

  • log in to PAWS
  • open a terminal
  • make sure you clone somewhere not public cd /tmp
  • run git clone
  • run the pip-command above
  • copy config cp
  • edit nano and add your credentials


Like my other tools, copy -> and enter the botusername (e.g. [email protected]) and password


It has 2 modes:

  1. automatic finding n-grams and trying to detect items that match (default if no arguments are given on the command line)
  2. add main subject items to scholarly articles

Both modes conclude by adding the validated or supplied QID to all scientific articles where the n-gram or search string appears (with spaces around it) in the label of the target item (e.g. scientific article).

Adding QIDs manually

Always provide the most precise subjects first

Run the script with the -l or -list argument followed by one or more QIDs:

  • python -l Q108528107

Here is a more advanced example: The first is metastatic breast cancer which is a subclass of the second breast cancer

  • python -l Q108528107 Q128581

In this case the tool is smart enough (thanks to Jan Ainali) to first add metastatic breast cancer to items and then exclude those items when adding the more general subject afterwards.

This way we avoid redundancy since we want the most specific subjects on the items and not all the general ones above it in the classification system.

Please investigate before adding broad subjects and try to nail down specific subjects and add them first. If you are unsure, please ask on-wiki or in the




GitHub - dpriskorn/ItemSubjector: Tool to add main subject to items on Wikidata using WMFs CirrusSearch for named entity recognition or a manually supplied list of QIDs
Tool to add main subject to items on Wikidata using WMFs CirrusSearch for named entity recognition or a manually supplied list of QIDs - GitHub - dpriskorn/ItemSubjector: Tool to add main subject t...