LINE-PESC/nlp-benchmarks

Keeping track of reported task results on common NLP tasks, using different datasets

nlp-benchmarks

Simple repo, where I want to keep track of reported task results on common NLP tasks, using different datasets

Related: ACL: State of the Art

Grouped by dataset/domain

• 20 Newsgroups

• BibSonomy ECML PKDD 2008

• BibSonomy ECML PKDD 2009

• CoNLL 2003 Named Entity Data (Sang and De Meulder 2003)

• Delicious Datasets

• IMDB Dataset (Maas et al. 2011)

• Movie-review-data: Sentence Polarity (Pang and Lee 2005)

• Reuters-21578

• RCV1-v2 (Lewis et al. 2004)

• Stackexchange Data: Kaggle 2013

• Stackoverflow Datasets

• Stanford Sentiment Treebank (Socher et al. 2013)