What is SkELL?

SkELL (Sketch Engine for Language Learning) is a simple tool for language learners and language teachers to easily check whether or how a particular phrase or a word is used by real users of the language.

desktop access
mobile version: SkELL mobile

Introduction to SkELL

See a brief presentation on using SkELL for English language teachers created by James Thomas or another introduction SkELL: Easy to use for teachers and students on the eflnotes blog created by Michael Houston Brown.

About SkELL

SkELL stands for Sketch Engine for Language Learning.

The Sketch Engine is a state-of-the-art web-based tool for building, managing and exploring large text collections in dozens of languages. It is used all over the world by many individuals, as well as companies such as Cambridge University Press, Oxford University Press and Macmillan. For more information, go to the Sketch Engine home page: http://www.sketchengine.co.uk/.

SkELL searches offer you three ways for exploring a cross-section of international varieties of English. The first is the concordance: for a given word or phrase, you will get up to 40 example sentences for any word or phrase you enter. The second is the word sketch through which you can discover typical collocates for your search word. And the third is the similar words (thesaurus) which lists words that are similar to, though not necessarily synonymous with, your search word. They are also visualised with a nice word cloud.

Data sources

SkELL is using a large text collection gathered specially for the purpose of the English language learning. It consists of texts from the news, academic papers, Wikipedia articles, open-source books, web pages, discussion forums, blogs, etc. There are more than 60 million sentences in the collection and more than one billion words which provide a sufficient coverage of both everyday, standard, formal and professional English language.


Concordance offers you a simple yet powerful search tool. It is similar to Google Search: if you type a word (e.g. controllable) or a phrase (e.g. in regard to) and click the Search button, you will get up to 40 example sentences featuring the query. SkELL is not for searching relevant documents as Google Search but it is for discovering how words behave in the English language.

Notice that the search is case-insensitive, i.e. you will obtain the same results for rutherford and Rutherford. Moreover, the results may contain the query (word or phrase) in a derived word form. If you search for mouse (the basic word form, lemma), it will find also sentences with mice. But if you try to search for mice you will get a different result (only mice words).

You do not need to specify part of speech (noun, verb, adjective, preposition, adverb, etc.), the search will find all of them by default. If you search for book, it will get sentences with book as a verb and as a noun and both in various word forms (booking, booked, books). You may narrow your search if you click on one of the suggested part-of-speech links, e.g. book as noun: book.


Word Sketch

Word sketch function is very useful for discovering collocates and for studying contextual behaviour of words. Collocates of a word are words which occur frequently together with the word – they co-locate with the word.

If you click on “Word Sketch”, you can search for collocates of a query (headword). E.g. for mouse you will get several tables which contain collocates of the headword mouse. The headers of the tables tell you what kind of collocates it contains.

By clicking on a collocate you will get a concordance with highlighted headwords and the collocate. This way you can see how the two words are usually used in English texts.

By default, the most frequent part of speech (noun, verb, adjective, etc.) is shown in Word Sketch. If a word (e.g. fast) can have more than one part of speech (in this case adjective, adverb, verb and noun), you can click on the alternative parts of speech above the tables.


See a screenshot of Word Sketch page optimised for mobile devices.

Similar words (thesaurus)

If you want to see words which are similar (not only synonyms) to your word (e.g. lunch) use the third function of SkELL: Similar words (thesaurus).

If you search for a single word, you will get a list of up to 40 most similar words visualised using word cloud. As in Word Sketch, if a word can have more than one part of speech, you can use links to the alternatives. When you click on a word you will be taken to the respective word sketch page for the word.


… more features

If you want to explore English and other languages in more detail:

use advanced concordance searches,
search for multiword word sketches,
explore bilingual corpora,
compare collocates of two headwords,
build your own corpora instantly,
generate word lists, n-gram lists etc.,
try Sketch Engine.

Check out a brief survey of some of many features of the Sketch Engine or Changelog listing all recently added features.


This service is provided by the copyright holders and contributors “as is” and any express or implied warranties, including, but not limited to, the implied warranties of merchantability and fitness for a particular purpose are disclaimed. In no event shall Lexical Computing Ltd. be liable for any direct, indirect, incidental, special, exemplary, or consequential damages (including, but not limited to, procurement of substitute goods or services; loss of use, data, or profits; or business interruption) however caused and on any theory of liability, whether in contract, strict liability, or tort (including negligence or otherwise) arising in any way out of the use of this software, even if advised of the possibility of such damage.

References to SkELL and versioning

From time to time, the underlying corpus data may be changed (cleaned, refined, …). If you are referring to particular results (using bookmarked URLs for example), you should refer also to a particular version. The web interface may also be changed occasionally. At the bottom of SkELL page, you can see a version of this format: VERSION1-VERSION2. The first corresponds to the version of the interface and the second with the version of the corpus data.

If you use results and examples from or links to SkELL, please, cite this paper:

BAISA, Vít a Vít SUCHOMEL. SkELL: Web Interface for English Language Learning. In Eighth Workshop on Recent Advances in Slavonic Natural Language Processing. Brno: Tribun EU, 2014, pp. 63-70, 8 p. ISSN 2336-4289.