What is the difference between these two searches

[mortenjensen]

Hi all,

I am trying to understand the difference between the following two search strings:

1: [RANGE mark] [COUNT 1- 1000000] @ [NOUN]

2: [RANGE mark] [NOUN] [COUNT 1-1000000]

 

The produce quite different results:

- The second marks pronouns.

- The first have 49 hits of "theos", the second only 37.

 

I want to find out which nouns are the most common in Mark. The first one gives the correct count, the second not. Why is that?

 

And just to complicate the matter, I tried this one too (which works, but I am not searching for what I want):

3: * [NOUN] [COUNT 1-100000] [RANGE Mark]

 

What am I in effect searching for here? If I changes it to this:* @ [NOUN] [COUNT 1-100000] [RANGE Mark], it gives the correct result again.

 

So I guess what I need to understand is the difference between and @-search and a search where you just type in commands in a line.

Thanks for clarifying and helping out!

 

Morten

[Ken Simpson]

Hi Morten, the differences between the two searches are this:

 

search 1

 

[Range Mark] = return the results from Mark

[Count 1-100000] = a hit is anything that occurs 1-100000 times in the text (all the NA28 in this case)

@[NOUN] = of the previous term (any word that occurs 1-100000 times) only “hit” the nouns. The @ character specifies that the succeeding value must apply to the preceding value)

 

search 2

[Range Mark]

[Count 1-100000] = see above

[NOUN] = a third term that is of any noun within the search scope (default a verse)

 

i.e. the first search is searching for 1 term - a word that occurs 1-100000 times that is also a noun

 

the second search is searching for 2 words, A noun,  followed by a word that occurs 1-100000 times (within the scope) - which is why it fails to find (for example) ΜΑΡΚΟΝ in Mark 1:0. There is no word that follows it.

 

As an aside, the [COUNT 1-100000] is redundant IMHO. If you drop [COUNT 1-100000]@ from the first search you get exactly the same result because there is no word that occurs more that 100000 times that it would exclude.

Edited April 29, 2015 by Ken Simpson

[mortenjensen]

Thanks Ken!

Now it makes sense.

If I want to find nouns only, my search string should be some like:

-  [RANGE mark] [noun]

- * @ [NOUN]  [RANGE mark]

 

If I want to find instances of a noun followed by any word (for which ever reason ), my search string will be:

- [RANGE mark] [NOUN] [COUNT 1-1000000]

 

If I want to find any word, followed by a noun, followed by any word, my search string will be:

-  * [NOUN] [COUNT 1-100000] [RANGE Mark]

 

The point is to remember that @ adds to the preceding tag, while to tags after one another just represent two word-searches after one another.

 

Thanks again,

 

Morten

[Joel Brown]

Morten, this is all correct.  The one tweak is that for COUNT type searches you can use + and - to determine endless bounds, like [COUNT 1+].

[Ken Simpson]

Yes Morten, though again, you don’t need the [COUNT 1-100000], unless you are constraining the search to a specific number, just use * it’s much faster to type! :-)