Kohlenberger/Mounce vs Analytics vs Even Shoshan Concordance

[Gedalya]

What accounts for the discrepencies between the K/M lexicon and the Accordance analytics of HTM-W4 and the Even Shoshan Concordance.

For eg, the name משה is listed as appearing 770 times in Even Shoshan but 767 times in K/M and 771 in the Accordance analytics

Other examples with verbs abound as well.

So how can I accurately know how many times a word appears in the Tanakh?

 

 

 

[Helen Brown]

I would suggest that Even Shoshan (the Hebrew concordance) and Accordance are close enough, so in K/M but if you want to know for sure, you do it the old fashioned way, by counting them!

 

For example, a word might get counted twice if it has variant parsing. K/M Dictionary has 4 entries for the same consonants, three Hebrew and one Aramaic. If you add the first two you get 770. If you analyse the search result in Accordance you get:

משׁה (771 total words)

Number of different forms = 4:
(Triple-click a form to see its occurrences)

        ‏משׁה‎    to draw = 3
        ‏(משׁה)    מֹשֶׁה    Moses = 766
        ‏(נשׁא)    מַשֶּׁה    loan = 1
        (משׁה)    מֹשֶׁה־0    Moses = 1

So all are correct, depending on which forms they include and exclude.

[Gedalya]

Thanks, Helen,

 

Why is   (משׁה)    מֹשֶׁה־0    Moses = 1" a separate entry from "(משׁה)    מֹשֶׁה    Moses = 766"

[Λύχνις Δαν]

The more I work with computer based text data the more I am convinced that one needs to know what decisions are embodied in the distillation that you work with at a user level. If you do the raw data stuff yourself then that's ok, you know what it is. If you don't questions like this arise frequently. I don't have a good solution to this problem until datasets themselves are fully documented which is a huge task for non-trivial databases.

 

Of course the counter-arguments are that most users mostly don't hit problems, which is mostly true. But then, this is the second example in a couple of days in this small part of the world, and it's certainly not isolated to Acc. This is a data issue.

 

Thx

D

[Ken Simpson]

Hi Gordon, the מֹשֶׁה־0 form in Ezra is Aramaic - that -0 notation delineates the Aramaic forms.

 

If you search for מֹשֶׁה and add the tag NOUN ProperName Accordance will find 767 occurrences.

 

Perfect with K/M - Even Shoshan I can’t help with sorry.

[Gedalya]

Ken,

Thanks very much.