building a corpus from COCA KWIC

Question

I'd like to look at an adjective + clause (adverbial, to-infinitival, -ing) combination in COCA and I'm happy with the KWIC results I got so far.

For my further analysis, it would be useful to have a corpus of my results with specific parts of the KWIC lines only. Is there a way to extract only those? I would be grateful for any suggestions.

What I have so far looks like this:

I would like to analyse my samples using AntConc - first create a corpus of all my samples and import it there. The problem is that a number of concordance lines overlap (e.g. when the sentence following the one with the bold parts would be "It is good to see you, too": In the COCA-results, each of these is counted only once but in a sub-corpus consisting of all the concordance lines, there would be duplicates).

So far, I've tried using Excel to copy the relevant part into a new cell with a function. In Excel, I also tried to split the cell with the line into several columns, choose the relevant columns and combine those again.

Answer 1

You can select the lines that you want to export, for example by clicking the little box on the top right next to the question mark in this screenshot:

As the warning message says you can apparently only export 100 per day; is that too much of a restriction? Else I suppose you can come back every day to export the next pages.