Viewing 4 reply threads
  • Author
    Posts
    • #47662
      Anastasios Asimakopoulos
      Keymaster
          @anastasios
          • Share the structure, the CQL sequence and the link to the results (concordance lines) below.
          • Reflect on the process of building your CQL e.g. How easy was it? Did you have to try a few times before you got it right? etc.
        • #53092
          Paula Acejo Cantero
          Participant
              @powla

              Hello,

              I focussed on the structure “verb happen followed by to be + pp” but I wanted to filter out regular participles just to try the function with exclamation mark.

              The query I used was [lemma=”happen”] [word=”to”] [tag=”VB.*”] [tag=”VVN.*” & word!=”.*ed”]

              I had to try several corpora, as I did not get enough results in BAWE or BASE, so I used BNC.

              I then used the frequency for the KWIC and selected the examples that I wanted (I did not include er, I thought it was not an usual verb)

              Concordance lines link

              The activity for the students could be to reflect on whether the word after to be has an adjectival function or a past participle one and how do they know the difference. The line between pp and ad can be very thin…

              How do we filter same KWIC results. I would prefer not to have happened to be born three times in my examples.

              Attachments:
              You must be logged in to view attached files.
              • #53121
                Anastasios Asimakopoulos
                Keymaster
                    @anastasios

                    Thank you for sharing your CQL @powla It is quite an advanced sequence since you combined tags, lemmas and various conditions (&, ! and .*ed), so well done! As you noticed, however, you didn’t get many results and that’s because the longer and more specific your search is, the fewer the results will be but that also depends on the size and type of corpus. A much larger, general corpus will return more results e.g. the structure occurs 6,367 times in enTenTen20 (43 billion words) with a relative frequency of 0.15 times per million words. Interestingly, happen + to be born accounts for 18.53% of the concordance lines and it is followed by happen + to be made (8.91%), set (6.52%), found (4.65%) and written (4.21%).

                    By the way, if you would like to exclude some concordance lines, you can tick the boxes on the left and then click NOT SELECTED (see below). This will exclude the selected concordance lines from your results.

                    Regarding the task itself, whether we have a case of the passive voice or an adjective, I think the sequence is limiting the data since we searched for VVN and excluded -ed forms. You could perhaps focus on VVN vs J and include words such as excited, interested, tired, etc.

                • #53117
                  Siti Asmiyah
                  Participant
                      @siti-asmiyah

                      My students often make mistake in using the word different. Influenced by our Bahasa Indonesia, my students often use the phrase different with instead of different from. Hence, I would like to have the structure related to the word different. I first tried to have examples from BAWE and BASE to find the possible structure. I then turn the possible structures in CQL and come up with the following structure.

                      [word=”different.* “] [word=”from.* “]

                       

                      Then I enlarged the search with the following structure

                      [tag=”NN.*”|tag=”NNS.*”] [tag=”VBZ.*” | tag=”VBR.*”] [word=”different.*”] [word=”from.*”]in BAWE and found the following list Concordance | Sketch Engine

                      I have tried to formula several times but then I found the feature of CQL Builder and work from that. I found this CQL Builder very handy.

                      Attachments:
                      You must be logged in to view attached files.
                      • #53124
                        Anastasios Asimakopoulos
                        Keymaster
                            @anastasios

                            Thank you for sharing your CQL @siti-asmiyah It is a great CQL search as it combines tags and words and the pipe symbol. To make it a little easier, I would recommend using the pipe symbol without retyping tag= – see below

                            [tag=”NN.*|NNS.*”] [tag=”VBZ.*|VBR.*”] [word=”different”] [word=”from”]

                            Also, you will notice that I removed .* from the words different and from. You don’t need to use .* with a word unless you are interested in a word family. For example, different.* included the example differentiated in your results. In addition, you might want to use [lemma=”be”] in order to include other forms of be e.g. was, were, been. Hope this helps. Keep up the good work!

                          • #53193
                            Siti Asmiyah
                            Participant
                                @siti-asmiyah

                                Thank you for the recommendation Anastasios

                            • #53227
                              Anastasios Asimakopoulos
                              Keymaster
                                  @anastasios

                                  You’re welcome @siti-asmiyah

                                • #53309
                                  Paula Acejo Cantero
                                  Participant
                                      @powla

                                      Thank you, very true, it was a bit too restructive the .ed parameter. I was actually more playing and trying to combine different types of SQL  :)

                                  Viewing 4 reply threads
                                  • You must be logged in to reply to this topic.