trWaC – Turkish corpus from the web

The Turkish Web Corpus (trWaC) is a Turkish corpus made up of texts collected from the Internet. The corpus was prepared according to standards described in the document A Corpus Factory for Many Languages (Kilgarriff et al. at LREC 2010). The corpus consists of 32 million words.

Part-of-speech tagset

The Turkish corpus was tagged by MaltParser with the pre-trained Turkish model. See this Turkish part-of-speech tagset.

Tools to work with the trWaC corpus

A complete set of tools is available to work with this Turkish corpus to generate:

  • word sketch – Turkish collocations categorized by grammatical relations
  • thesaurus – synonyms and similar words for every word
  • word lists – lists of Turkish nouns, verbs, adjectives etc. organized by frequency
  • n-grams – frequency list of multi-word units
  • concordance – examples in context

Changelog

version 3.0 (19 Nov 2014)

  • added document structures
  • added lempos
  • corrected tokenisation of quotes

version 2.0 (25 Oct 2011)

  • word sketches are compiled using below resources.
    • The morphological analyser and morphological disambiguator (POS tagger) from Kemal Oflazer and Deniz Yüret downloadable at  http://deniz.yuret.com/turkish/tr-disamb.tgz.
    • generated from an existing dependency parser. Dependency parser can be downloaded from http://web.itu.edu.tr/gulsenc/TurkishDepModel.html
    • thanks to Gülşen Eryiğit and Kemal Oflazer for answering our emails and providing us with the tools

Bibliography

BARONI, Marco, et al. The WaCky wide web: a collection of very large linguistically processed web-crawled corporaLanguage resources and evaluation, 2009, 43.3: 209-226.

Corpus factory method

Adam Kilgarriff, Siva Reddy, Jan Pomikálek, and Avinesh PVS. A corpus factory for many languages. In LREC workshop on Web Services and Processing Pipelines, Malta, May 2010.

Turkish Word sketches 

Ambati, Bharat Ram, Siva Reddy, and Adam Kilgarriff (2012). Word Sketches for Turkish. In Proceedings of the Eight International Conference on Language Resources and Evaluation (LREC’12), pp. 2945–2950.

Search the LANG corpus

Sketch Engine offers a range of tools to work with this LANG corpus from the web.

or

Other text corpora

Sketch Engine offers 450+ language corpora.

Use Sketch Engine in minutes

Generate collocations, frequency lists, examples in contexts, n-grams or extract terms. Use our Quick Start Guide to learn it in minutes.