The ATA Chronicle - September/October 2017 - 36

RESOURCE REVIEW |

BY JOST ZETZSCHE

Language Sketching

rago Ciobanu, from the translation
department of the University of
Leeds, wrote to me a few weeks ago:

The reason I'm writing is to ask
whether you've had a chance to play
with Sketch Engine (see sketchengine.
co.uk). In Leeds, we've been using it in
our corpus linguistics work a lot. It's got
brilliant features, from the terabytes of
super useful multilingual data which it
already comes with, to features for term
extraction, specialized corpus building,
thesaurus, collocations, and tons more!
It's really, really cool and I'm only
writing to you because the translators
I know who have been playing with it
also like it a lot.

Not sure whether you could tell, but
Drago really likes Sketch Engine. And in

a way, I could stop this column right here,
because he already said it all-sort of.
After spending some time looking at
Sketch Engine, I felt embarrassed that
I hadn't known more about it. As
Drago said, it's really, really cool. It's also
a monster of a tool (size-wise) and it's not
particularly easy to navigate when you
first encounter it. (According to
Ond ej Matuška of the Sketch Engine
team, one of the areas they're trying to
focus on in the immediate future is to
make the product more user-friendly.)
But first, what exactly is Sketch Engine
and what does it do?
It's a corpus tool developed by the
Czech company Lexical Computing
Limited. Lexical Computing was originally
founded in 2003 by the late Brit Adam
Kilgarriff and Pavel Rychlý, a professor

at Masaryk University in Brno. The idea
of corpus tools, and this corpus tool in
particular, is to find how language behaves
based on large collections of data. For this
purpose, Sketch Engine built corpora in
more than 80 languages (as well as "timestamped" corpora in a slightly different
set of 18 languages for the purpose of
comparing word usage over time). The
sizes of the corpora differ widely (from
just a few million words in Maori to more
than 800 billion in English), and they
are available for a number of analysis
purposes for any paying trial user. (The
annual subscription price is 100 euros for
non-academic users, with the trial period
ending after 30 days.)
The analyses you can do on these corpora
with Sketch Engine include the following:

Figure 1: An example of a Word Sketch
Remember, if you have any ideas and/or suggestions regarding helpful resources or tools you would like to see featured, please e-mail me at jzetzsche@internationalwriters.com.

The ATA Chronicle | September/October 2017

www.atanet.org

http://www.sketchengine.co.uk http://www.sketchengine.co.uk http://www.atanet.org

Table of Contents for the Digital Edition of The ATA Chronicle - September/October 2017