CQL 1: Complex corpus searches with the Corpus Query Language

[@andreaantal9660]

This was very useful! Thanks!

[@SketchEngine]

Thank you. We are happy to hear you liked it.

[@lovebts774]

I am trying to search for all sentences including "make + someone/something/nouns + Verb" and I am using this formula: [lemma = "make" ] [ ] {0,10}[tag = V.*] but I cannot find the result. Can you help me fix it? Thank you. By the way, {1.10] means any words in between 'MAKE' and ' a verb' is ok

[@charlienbredefeld4392]

I am wondering if it is possible to define a noun [tag="N.*"] that is written starting with a capital letter (I am looking at a learner corpus here). As Sketch Engine only has the uncoded corpus available, that would be my attempt to find the errors on capitilazation in English writing of learners. Also, is there a way to exclude names from the results? I could not find a tag for names.

[@SketchEngine]

This is the CQL you should use [word="[[:upper:]].*" & tag="N.*"]

[@charlienbredefeld4392]

@@SketchEngine Thanks a lot!! Is there also a way to exclude words at the start of a sentence?

[@SketchEngine]

@@charlienbredefeld4392 There should be but I do not quite understand what you mean. Can you give a concrete example of what exactly you want to find?

[@charlienbredefeld4392]

​@@SketchEngine Sure, I want to find all falsely capitalized nouns in the learner corpus. Therefore I need to exclude all proper nouns from the search and all nouns that stand at the beginning of a sentence. This is how far I have come: [word="[[:upper:]].*" & tag="N.*" & tag!="NP"] I am missing the code for "not at the beginning of the sentence" Thanks so much.

[@SketchEngine]

Not completely easy. We are thinking.....

[@laurac7269]

Hello! I have a question: I'm looking for N + PREP. + pierre and I'm using the cql: [tag="NOM"] [tag="PRE.*"] [lemma="pierre"] , but it's not working. Could someone help me please? PS [tag="NOM"] [ ] [lemma="pierre"] is actually working!

[@ondrejmatuska8945]

Are you sure the tagset of your corpus contains the tag PREP? Please generate the result (even the empty incorrect result) in the interface and click the "Request help and support" icon in top right of the interface and type your question. This way our support team will receive the necessary additional information to be able to help.

[@laurac7269]

@@ondrejmatuska8945 I gave a look to the tagset and solved the problem (it’s “PRP” and not “PRE” as I originally typed ). Thanks :)

[@laurac7269]

@@ondrejmatuska8945 there is a mistake in the values Sketch Engine. The right tag for prepositions is PRP! I’ve reported it.

[@ondrejmatuska8945]

@@laurac7269 When working with tags, it is always useful to refger to the corpus infor page which shows the tags for the corpus. The tagsets may differ between corpora, especially if you use some less common corpora such as historical corpora.

[@emmajuul4592]

Hi. I am working on this search: [word="som"][tag="P.*"][tag="N.*"][tag="P.*"][lemma="være"] (in Danish, but I don't think it matters for my question). How do I exclude examples/results, where a new sentence or paragraph begins after the noun? I am searching for results without a full stop "." between the noun and the pronoun. If my question needs to be specified, please say so. Thank you in advance :-)

[@SketchEngine]

You should use the within operator as explained in another video here: kzbin.info/www/bejne/nWisqK2Ng9-bd6s

[@atikhomthienthong416]

I want to identify passive structures. How should I do?

[@SketchEngine]

Sorry for the late reply. There is no magic button to find passive structures, you have to use CQL to search for be + -ed or similar. Please email [e-mail] if you need more help.