Morphological Search Quirks/BUGS

Dave Hooton

This shouldn't be one's introduction to Morphological Searching. What on earth is Andersen-Forbes Aram..?

This is not much better ...

Suggest that something more intuitive than obscure types of morphology confront the (average) user:-

===============

Poor man's Genitive Absolute (using above NA27):

@V??P??G BEFORE 3 words @NG produces 342 in 162 verses

The Morph dialog box hindered trying to type @JG as an OR'd term:-

@V??P??G BEFORE 3 words (@NG, @JG)

But the result was ZERO.

BUG: This is a legal query in v3 Morph Search (and see last query)

(@V??P??G BEFORE 3 words @NG) OR (@V??P??G BEFORE 3 words @JG) succeeded with 426 in 187 verses.

I tried a query with OR'd text terms:-

@V??P??G BEFORE 3 words (false, many) produced 23 in 9 verses.

BUG: the various results include glosses like much, more irrespective of "Match all word forms".

Find more posts tagged with

Features

Comments

Dave Hooton

Note results of @V??P??G BEFORE 3 words (@NG, @JG) only occur in NASB95 --- Beta 4

a. Result not shown

OK - I truncated the space that was no result!

b. Result and Bible

Robert Pavich

https://community.logos.com/discussion/comment/15856#Comment_15856

Dave,

what's wrong with this picture? I cut/pasted your query and look at the results.

Dave Hooton

https://community.logos.com/discussion/comment/15876#Comment_15876

The only thing wrong are the pictures. Check this:

Note the lack of results in the header of the bible and the different results presentation??

Your picture shows the format of Basic Search results! I'm running Beta 4 4.0.2.9821

Robert Pavich

https://community.logos.com/discussion/comment/15901#Comment_15901

My Cambridge paragraph bible must be much better than yours!

Dave Hooton

https://community.logos.com/discussion/comment/15903#Comment_15903

My Cambridge paragraph bible must be much better than yours!

Bob, look closer

Your picture shows the format of Basic Search results! I'm running Beta 4 4.0.2.9821

Rick Brannan (Logos)

https://community.logos.com/discussion/comment/15905#Comment_15905

Your picture shows the format of Basic Search results!

Hi Dave.

Just ran this (all passages, all resources, Logos Greek Morphology, beta 4 with all resource updates, 12,930 results in 4,022 articles, 1.20 sec) and get the library results as well. This is a guess on my part, but perhaps Robert and I are returning non-Bible hits (e.g. Apostolic Fathers, Philo, Josephus) as well as Bible hits, hence the library results, while you're just returning Bible hits (NT/LXX based) hence the morph results? When I limit the search to an NT (NA27, 426 results in 187 verses, 0.69 sec) I get the 'verses" results.

Dave Hooton

https://community.logos.com/discussion/comment/15907#Comment_15907

This is a guess on my part, but perhaps Robert and I are returning non-Bible hits (e.g. Apostolic Fathers, Philo, Josephus) as well as Bible hits, hence the library results, while you're just returning Bible hits (NT/LXX based) hence the morph results?

Rick, thanks for your quick response. What you say is true as I don't have those non-bibles. The only way I get your result in NA27 is to use the long form:

(@V??P??G BEFORE 3 words @NG) OR (@V??P??G BEFORE 3 words @JG)

Yet you got your results with @V??P??G BEFORE 3 words (@NG, @JG) ?

It would be good to have Bradley clear up this issue.

Rick Brannan (Logos)

https://community.logos.com/discussion/comment/15909#Comment_15909

Rick, thanks for your quick response. What you say is true as I don't have those non-bibles. The only way I get your result in NA27 is to use the long form:

(@V??P??G BEFORE 3 words @NG) OR (@V??P??G BEFORE 3 words @JG)

Yet you got your results with @V??P??G BEFORE 3 words (@NG, @JG) ?

Hi Dave.

I used what you're calling the "long form".

Dave Hooton

https://community.logos.com/discussion/comment/15910#Comment_15910

I used what you're calling the "long form".

There's still an issue as to why the "short form" works as per Bob's results!

I should say, "appears to work"!

davidphillips

https://community.logos.com/discussion/comment/15912#Comment_15912

I should say, "appears to work"!

This would be a correct way to phrase it. I also get results similar to Bob and Rick, but the results are not accurate. For example, the query returns a result in Matthew 22:2, and "heaven" is highlighted. "Heaven" is a genitive noun, but there is not a genitive participle anywhere nearby. It seems like the search is being read as follows:

@V??P??G OR @NG OR @JG

Well, that's not quite right. I ran that search and compared the results in the Cambridge Paragraph Bible. They are as follows:

@V??P??G BEFORE 3 words (@NG, @JG) - 3597 results

@V??P??G OR @NG OR @ JG - 3845 results

What it isn't doing is the actual long form of the search:

(@V??P??G BEFORE 3 words @ NG) OR (@V??P??G BEFORE 3 words @JG) returns only 251 results in the Cambridge English Bible.

The phrase (@NG, @JG) syntax is being read, but not correctly, by English language resources, while it is not being read at all by logos morphology resources.

But there's more!

@V??P??G BEFORE 3 words (@NG, @JG) is read by Gramcord resources - it returns 111 results in the NA27 int. But it returns 0 results in the NA27. It isn't returning the intended query, however. In fact, the results don't really make sense at all. The "long form" of the search returns 742 results in the NA27 and 742 results in the NA27 int.

Dave Hooton

https://community.logos.com/discussion/comment/16205#Comment_16205

I should say, "appears to work"!

This would be a correct way to phrase it. I also get results similar to Bob and Rick, but the results are not accurate.

Yes, I knew that when I saw the numbers. It is one of the problems with Morphological Search. Consider this:

In Morph Search: type & select εἰμί + type and build morph VPAI3S, which results in:

lemma:εἰμί @VPAI3S

Type operator ANDEQUAL and Search:

lemma:εἰμί ANDEQUAL @VPAI3S ==> Zero results

Now transfer to Bible Search ==> (<lemma = lbs/el/εἰμί> AND "ANDEQUAL" AND <lbs-morph+el ~ VPAI3S??>)

Correct and Search:

<lemma = lbs/el/εἰμί> ANDEQUAL <lbs-morph+el ~ VPAI3S??> ==> 897 in 815 verses in NA27.

Now, we can do this with Morph Search in v3 and get the same results! L4 Morph Search does not recognise ANDEQUAL so we have to transfer it to Bible Search.

Dave Hooton

https://community.logos.com/discussion/comment/16205#Comment_16205

But there's more!
@V??P??G BEFORE 3 words (@NG, @JG) is read by Gramcord resources - it returns 111 results in the NA27 int. But it returns 0 results in the NA27. It isn't returning the intended query, however. In fact, the results don't really make sense at all. The "long form" of the search returns 742 results in the NA27 and 742 results in the NA27 int.

The Morphology is incorrect for GRAMCORD. For (Verb, Participle, Genitive) BEFORE 3 words ( Noun, Genitive OR Adjective, Genitive) ==>

@V????P??G BEFORE 3 words (@N??G, @J??G)

And we end up with the same issues!

davidphillips

https://community.logos.com/discussion/comment/16497#Comment_16497

Shame on me. Good catch, Dave. I cut and paste without thinking about it.

Dave Hooton

https://community.logos.com/discussion/comment/16778#Comment_16778

Shame on me.

We all do it .. no shame!

I appreciate your contribution.

Dave Hooton

https://community.logos.com/discussion/comment/16996#Comment_16996

When will we see some improvements to the above?

Bradley Grainger (Logos)

https://community.logos.com/discussion/comment/18130#Comment_18130

When will we see some improvements to the above?

Probably following Beta 6; I haven't been able to work through the issues you've raised, but have flagged this thread for follow-up.