why the duplicates in Cascadia?

Page 1 of 2 (22 items) 1 2 Next >
This post has 21 Replies | 1 Follower

Posts 4508
Forum MVP
Robert Pavich | Forum Activity | Posted: Thu, Dec 31 2009 4:01 AM

I tried to duplicate a search that I've created in OpenText Graphs, in Cascadia...and whew! I finally got it to work

BUT....

It shows duplicate results???

Can someone explain why? Did I do something wrong?

See the comparison screen shot with Opentext...it found the same hits...but OpenText only reported them once....

Robert Pavich

For help go to the Wiki: http://wiki.logos.com/Table_of_Contents__

Posts 5337
Forum MVP
Kevin Becker | Forum Activity | Replied: Thu, Dec 31 2009 4:06 AM

Did you match skipped levels in the Cascadia and not the Opentext? (if not, could you post a picture of your query so I could duplicate it?) That seems like the most likely culprit.

Posts 4508
Forum MVP
Robert Pavich | Forum Activity | Replied: Thu, Dec 31 2009 4:58 AM

Kevin...

You were right...it was my "willy-nilly-matching-skips-levels" that did it....

I got it to be accurate (at least match the OpenText resource) BUT now..I have a Q:

 

How do I get it to include John19:12 (as the OT does)

 

 

Robert Pavich

For help go to the Wiki: http://wiki.logos.com/Table_of_Contents__

Posts 5337
Forum MVP
Kevin Becker | Forum Activity | Replied: Thu, Dec 31 2009 6:01 AM

This included John 19:12

EDIT: And as I look at the results the Unordered group is not needed, but it's hard to delete without everything attached to it below... that would be a feature I would like, delete only this node and attach its children to its parent.

Posts 4508
Forum MVP
Robert Pavich | Forum Activity | Replied: Thu, Dec 31 2009 6:09 AM

Kevin,

I just did that search and now I'm back to 400 results because of dupes!

Which nodes got "matching skips levels?

Robert Pavich

For help go to the Wiki: http://wiki.logos.com/Table_of_Contents__

Posts 5337
Forum MVP
Kevin Becker | Forum Activity | Replied: Thu, Dec 31 2009 6:12 AM

I only matched skipped levels to the final terminal node, I did this because there were some clause functions intervening (you can tell by the dashed lines between nodes, solid lines indicate direct parent child relationship). I'll go back to the drawing board and see if I can get a query with fewer duplicates.

Posts 4508
Forum MVP
Robert Pavich | Forum Activity | Replied: Thu, Dec 31 2009 6:15 AM

Kevin,

that's what I have; only the terminal nodes skip levels but I'm still getting 391 repeat hits...

Robert Pavich

For help go to the Wiki: http://wiki.logos.com/Table_of_Contents__

Posts 5337
Forum MVP
Kevin Becker | Forum Activity | Replied: Thu, Dec 31 2009 6:17 AM

This was trimmed down to 225 and is a little more elegant.

Posts 4508
Forum MVP
Robert Pavich | Forum Activity | Replied: Thu, Dec 31 2009 6:21 AM

Kevin Becker:
This was trimmed down to 225 and is a little more elegant.

 

lol...that's like killing spiders with a 12 gage!.....

 

I'm looking to just catch the last couple of strays....that's' all Big Smile

Robert Pavich

For help go to the Wiki: http://wiki.logos.com/Table_of_Contents__

Posts 4508
Forum MVP
Robert Pavich | Forum Activity | Replied: Thu, Dec 31 2009 6:37 AM

On your last "refined" version; I searched john only and got 36 results; John 19 was included but there were duplicates again.....I'm not understanding why?

Robert Pavich

For help go to the Wiki: http://wiki.logos.com/Table_of_Contents__

Posts 5337
Forum MVP
Kevin Becker | Forum Activity | Replied: Thu, Dec 31 2009 6:38 AM

are shotguns allowed on the forums? Wink

I look forward to verse lists coming back, they will help in evaluating the results of a syntax search. At least in searches with high amounts of duplicates they don't duplicate references.

However, I trimmed the search down to the "branch" with the match skipped levels in it. It came back with only 77 results and not that many duplicates (it accounted for all but 3 of the duplicates in the 225 search) . Could this syntax structure you are looking for be common? Or could the participle in the John 19:12 not really fit? I ask because I don't understand how this information is useful... it's a good thing to understand why someone is looking for something Big Smile.

Posts 5337
Forum MVP
Kevin Becker | Forum Activity | Replied: Thu, Dec 31 2009 6:41 AM

Robert Pavich:
On your last "refined" version; I searched john only and got 36 results; John 19 was included but there were duplicates again.....I'm not understanding why?

You did something different than me. I didn't get these duplicates. Did you take out the match skipped levels in the last branch?

 

Posts 4508
Forum MVP
Robert Pavich | Forum Activity | Replied: Thu, Dec 31 2009 6:50 AM

Kevin,

I guess I might have done something different than you did....I'll go back over it and check.

 

I'm doing this search (actually redoing it) because I'm trying to see if the phrase "all the ones who do x" or "all the ones who are x" is something that John uses frequently....and it seems he does.

I'm doing it because of the rampant belief in John 3:16 that the same phrase carries a LOAD of excess baggage beyond the "all the believing ones" not perish but gain eternal life.

Do you know how people turn that statement into more than is there? (does that makes sense?)

That was the reason to start this search in OpenText in v3 and I was just trying to do two things:

1.) Recreate it in V4

2.) Practice with Cascadia with something that is somewhat familiar to me.

Robert Pavich

For help go to the Wiki: http://wiki.logos.com/Table_of_Contents__

Posts 4508
Forum MVP
Robert Pavich | Forum Activity | Replied: Thu, Dec 31 2009 6:56 AM

Ok....

I redid your last search two ways:

  1. ONLY the terminal nodes skipping levels....
  2. ONLY the terminal nodes skipping levels except for the one under the "clause" that you added last.

 

In case #1 I get 38 hits and dupes as you saw.

In case #2 I get the same results as not having that extra branch....I miss out on John 19.

Robert Pavich

For help go to the Wiki: http://wiki.logos.com/Table_of_Contents__

Posts 5337
Forum MVP
Kevin Becker | Forum Activity | Replied: Thu, Dec 31 2009 6:56 AM

Robert Pavich:
Do you know how people turn that statement into more than is there? (does that makes sense?)

It makes sense, but off the top of my head I can't remember any specific example for this verse. It seems to me that in both Open Text and Cascadia you have to struggle to find a happy medium, skipping levels gives you the best chance to catch variation, but it gives you a lot of duplicate hits to sort through, but clamping down on the duplicates runs the risk of eliminating a legitimate hit.

And for a final throw away inquiry: would a Morph proximity search have been better for this question?

Posts 4508
Forum MVP
Robert Pavich | Forum Activity | Replied: Thu, Dec 31 2009 7:01 AM

Kevin Becker:
And for a final throw away inquiry: would a Morph proximity search have been better for this question?

Kevin,

actually, I don't know. As I said..this came about because of my previous search....If the syntax search had proximity operators that would be great...lol..

 

And you are right...the balance between "loose enough and tight enough" is a hard one to come up with...

 

can you help me on a morph search for this construction?

Robert Pavich

For help go to the Wiki: http://wiki.logos.com/Table_of_Contents__

Posts 5337
Forum MVP
Kevin Becker | Forum Activity | Replied: Thu, Dec 31 2009 7:05 AM

Robert Pavich:

I redid your last search two ways:

  1. ONLY the terminal nodes skipping levels....
  2. ONLY the terminal nodes skipping levels except for the one under the "clause" that you added last.

 

In case #1 I get 38 hits and dupes as you saw.

In case #2 I get the same results as not having that extra branch....I miss out on John 19.

This is odd, we're getting our signals cross some way (I can't wait until they allow us to share user files to make this sort of collaboration easier). Here's my inquiry again, it catches John 19 for me.

Regarding the Morph search, sure, I'll start playing around to see if I can get a satisfactory one.

Posts 4508
Forum MVP
Robert Pavich | Forum Activity | Replied: Thu, Dec 31 2009 7:21 AM

Kevin,

I think that on that last search, it was me who hosed things up....

I carefully went back and re-created your last search from scratch and got 24 nice, relevant hits from John!

I'd still like to see if a morph search could have done this....but this will do for my syntax search needs....

 

Thank you very much for your patience!

 

Robert Pavich

For help go to the Wiki: http://wiki.logos.com/Table_of_Contents__

Posts 5337
Forum MVP
Kevin Becker | Forum Activity | Replied: Thu, Dec 31 2009 7:35 AM

You're welcome.

I ran this query

(lemma:πᾶς WITHIN 2 WORDS @D WITHIN 3 WORDS @V??P) OR (lemma:πᾶς WITHIN 1 WORDS @RR WITHIN 2 WORD @V)

I don't think it's a substitute for the syntax search and it does have at least one bogus hit, but it catches some things I'd want to account for with your question that the syntax search doesn't account for. Looking at one of its "new" hits I see why Cascadia missed it. I'm not certain I have the energy to modify the search to make it work, but I think the lesson I take from this is that a morph proximity search can help you find syntax examples that your initial approach missed.

Posts 4508
Forum MVP
Robert Pavich | Forum Activity | Replied: Thu, Dec 31 2009 7:54 AM

Kevin,

thanks very much....

It's too bad that I spent so long on the syntax search and it was actually easier (because of what I was looking for) to do it via morph search....

I'll play around with this....thanks very much again... Big Smile

Robert Pavich

For help go to the Wiki: http://wiki.logos.com/Table_of_Contents__

Page 1 of 2 (22 items) 1 2 Next > | RSS