Josef Ruppenhofer

Datasets

Modality

  • Modals in the MPQA corpus

    For the LREC-conference 2012, Ines Rehbein and I created word-sense and frame role annotations for the instances of 5 modal verbs (may/might, must, shall, ought, can/could) in the MPQA (multi-perspective question answering corpus). Get the data here.

Frame semantic annotations

  • Salsa Frame-semantic lexicon and annotations for German

    As a post-doc I was involved in the second phase of Salsa, a German sister project to FrameNet.

Semeval 2010 Shared Task on Null Instantiation

  • Goldstandard data for the Semeval 2010 Shared Task
  • This data used to also be available on pages at the dept of computational linguistics (CoLi) at Saarland University but seems to have been taken down. Find it linked below.
If you only want to look at the data, you can do so at the FrameNet website. The Shared Task data has been incorporated into FrameNet’s annotated corpus.

Sentiment

  • MPQA Sentiment annotations for English-language news and press data

    As a post-doc I was involved in the second phase of MPQA.

  • MLSA Corpus A Multi-Layered Reference Corpus for German Sentiment Analysis created by the IGGSA group.

  • Opinion Role Lexica and Corpus Annotation Opinion-role resources created by my project partner Michael Wiegand (Saarland University), related to this publication .

  • German EffektGermaNet Effect annotations for German synsets contained in GermaNet 9.0, related to this WASSA paper .

GermEval Shared Tasks on Offensive Language

  • Germeval2018 Shared Task on the Identification of Offensive Language. “https://github.com/uds-lsv/GermEval-2018-Data”>data]

  • Germeval2019 Shared Task on the Identification of Offensive Language. [data]

  • Many further resources related to collaborations with Michael Wiegand around Sentiment and Offensive Language are available at his resource page.