Graded Word Meaning Annotation

• Data and Guidelines:

  Second Round 2013

  Katrin Erk, Diana McCarthy and Nick Gaylord (2013) Measuring Word Meaning in Context. Computational Linguistics 39 (3) pp 511-554 DOI 10.1162/COLI_a_00142 article available here

  Download the data here Do send us an email to diana at dianamccarthy.co.uk if use this data

  Guidelines given to the annotators:
  • Word Sense Similarity (WSsim-2) Guidelines graded annotation of WordNet senses on a 5-point scale
  • Usage Similarity (Usim-2) Guidelines graded annotation of usage similarity on a 5-point scale
  • Word Sense Best (WSbest) Guidelines traditional single-sense annotation
  • Synonym Best (SYNbest) Guidelines lexical substitution

  First Round 2009

  Katrin Erk, Diana McCarthy and Nick Gaylord (2009) Investigations on Word Senses and Word Usages In Proceedings of the Joint conference of the 47th Annual Meeting of the Association for Computational Linguistics and the 4th International Joint Conference on Natural Language Processing of the Asian Federation of Natural Language Processing ACL-IJCNLP Singapore

  The Data and Annotator Instructions for this 2009 paper are available from Katrin's web site 2009 dataset and guidelines here
  NB this earlier data is referred to as the 'first round' in the 2012 Computational Linguistics paper.

  Related work:

  See also Dominik Schlechtweg's Word Usage Graphs. These are graphs of usage similarity. He has also provided a script for transforming a dataset of usage similarity data, such as ours, into word usage graphs.