View Full Version : filtering results of an IEL search
02-11-2009, 11:42 AM
I need to find all participles of words which occur 50x or more in the New Testament, and I am blanking out on how to do this.
What I have done is created an IEL with all the verbs that occur 50x or more in the BNM, and I can readily find those. What I can't figure out is how to use the agreement box in the GSE to limit the finds to participles only.
Any and all help is appreciated!
02-11-2009, 12:32 PM
Don't use the agreement box. Instead, open up that iel in the Wordlist Manager (WLM).
When you compiled it, did you do it without Greek accents? If not, recompile.
Then, and here's the fun part, select the whole list. Click on Edit>Change Morphology on Selected Words and type in "@np*" without quotes.
Save that as an iel list. Now open up the GSE, and load that IEL file.
Should do the trick.
(Note: I think there are a few ways to do this, but minor bugs prevented me from doing it other ways. There are selection issues in the WLM. If you try to select multiple items by selecting the word on one end, and then shift-clicking the word on the other end, it works. But only if you choose the one on top first. If you go the other way, it doesn't work.)
Here's what I would do, starting from scratch:
Open up the word list manager
With focus on Main Word List, click the Load/Generate Word List
Select BNM, Load words from a Bible version, uncheck Use Search Window limits (in case you had a limit on), Filter words using: *@v*, uncheck Keep morph codes, uncheck Keep Greek accents > Create list
You have now created a list of all verbs; Sort by frequency (from menu); scroll down until you find the first word with less than 50; click on it then hit CTRL-SHFT-End to highlight to the bottom of the list > then use Edit > Delete selected >> Here's your list of 50+ verbs in the NT [This should be the list that you already created and saved as an IEL. Once you have this list, you could have loaded that in and saved steps 1-4)
Now click on the Secondary Word List to change focus and click the Load/Generate Word List
Use same parameters as in step 3 EXCEPT for Filter words using put: *@vp* > Create list
In the secondary list, you now have a list of all participles in NT
Use Select > Select words common to both lists
The highlighted words are all the participles based on verbs used 50+ times in NT with the # of times participial forms are used. You don't need to use the GSE at all.
If you want to save that list of verbs, use Select > Invert Selection and then Edit > Delete selected, then save the list.
BTW, I ended up with 88 words occurring 3772 times that match your request.
02-11-2009, 05:52 PM
Thanks for the responses. These suggestions seem counterintuitive to me, though, because of the way I work. At DTS (as most everywhere) in our Greek classes we require students to memorize words which occur 50x or more in the NT. So this list of words will be used frequently in teaching situations. I'd much rather have the list of words saved as an IEL which I can then filter on the spur of the moment in the GSE.
I would guess from what you have written that there's no way to filter the results of an IEL in the GSE?
02-11-2009, 05:59 PM
You can always recompile the results of the search I did in the WLM and then save them as an IEL.
The first way I gave only returned the words that matched your request. You would have had to work through them one at a time to extract the verses in which they occur.
If you want the verses, then you should use the GSE
you will want to use the handy 50plus list that you created.
I've attached the Query you need and the 50plus times list that I created. Both of them should be saved to the ASE subdirectory. (You can use your own list, but you will need to change it in the GSE.)
In the GSE:
In the first word box, double-click to open > select correct version (BNM), click on Inclusion/exclusion list > Load IEL and load up my/your 50plus times word list.
Add another word box, double-click to open, select BNM, click on Word and enter * in the word box and @vp* in the Morph box. (This is the VerbParticipleAnything code. Change this code as desired for your desired search.) Click OK
Now add an Agreement box.
Now use Word Connect tool and connect the two word boxes to the AND. Then connect the two word boxes to the Agreement box.
Now double-click the Agreement box. Check the Lemma box and the Entire morphology box. (This is going to force the words in the two word boxes to be the same lemma.) OK
Run your search.
Running the search this way, I get 2897 verses with 3856 hits. Using the Word List Manager method, I had 3772 hits. I'm not exactly sure why there is a difference.
In any case, you should be able to use your 50 plus list and simply change the morph code in the second word box to adjust your desired results on the fly.
02-12-2009, 10:08 AM
Here is another way that may suit you better. (It requires BW8.)
Create your 50x word list using the WLM. Be sure that morph codes and accents are unchecked when you create the list. Save the list as an IEL. Be sure that when you name the list you do not include any spaces in the name, or it won't work. I also suggest including the name of the version used to the create the list. For example, BNM50xverbs.iel.
Now go to the Command Line. Set your search version to BNM (assuming the IEL is from BNM). Enter a period, then right-click on the Command Line and select "Insert Word List File Name" from the context menu. Enter the name of the 50x IEL.
The filename appears on the command line in angle brackets, like so:
Now add the @ sign, followed by the codes for the participle "vp". Be sure to add the asterisk at the end of the code.
The resultant search is:
When run, the search returns all the words in the word list that appear in the text as participles. I found 2897 verses, 949 forms, and 3772 hits. Of the 89 original search words, all except one (FHMI) have participial forms in the NT.
Hey, that is outstanding, Glenn! What a neat trick. Thanks for pointing out this new BW8 feature!
02-12-2009, 01:49 PM
The same thing works with Louw-Nida semantic domains, too. Use the Command Line option to insert the domain, then apply the @ plus the desired codes. In this way you can search a domain to pull out all the verbs, or all the nouns, etc. in that domain.
Powered by vBulletin® Version 4.2.2 Copyright © 2014 vBulletin Solutions, Inc. All rights reserved.