Computational Linguistics & Phonetics Computational Linguistics & Phonetics Fachrichtung 4.7 Universität des Saarlandes

SALSA II - The Saarbrücken Lexical Semantics Acquisition Project

SALSA The aim of SALSA II is to
  • provide a large, frame-based lexicon for German, with rich semantic and syntactic properties, as a resource for linguistic and computational linguistic research,
  • investigate probabilistic and hybrid methods for wide-coverage semantic annotation, and
  • explore the use of frame semantic annotations for dynamic semantic analysis in practical NLP applications, especially information access.
For more information, see our overview page.  

SALSA II is funded by the German Science Foundation DFG and builds on its predecessor project SALSA, in particular the German SALSA/TIGER corpus with manually annotated FrameNet frame structures.

Open Positions

No open positions at the moment.


  • MaJo - a toolkit for supervised Word Sense Disambiguation and Active Learning
    is now available.
  • For a detailed description of the toolkit please refer to our TLT-8 paper.
  • FrameNet Transformer - a tool for deriving custom versions of the FrameNet database release is now available
  • Explore the Salsa lexicon (version 1.0)
  • Search the Salsa corpus online here; for a brief overview see here, for information about the query syntax please refer to Torsten Marek's Master Thesis (.pdf) or get a quick overview here.
  • PATH SEARCH THROUGH FRAMENET release v 1.0 now available here.
  • FATE corpus release v 0.9 now available here.
  • Second SALSA Release now available here.
    (You can still download SALSA 1.0 from the same URL)