keyword analysis - log-likelihood values too high?

[Eugenia Diegoli]

Hi Laurence (if I may), hi everyone,

I'm using AntConc to identify the keywords in two specialised web corpora of Japanese (ca.1.5 million tokens each), using the 2020 100k sentence Leipzig web corpus as reference; I ordered the keywords by log-likelihood and set a minimum range value of 500. The LL values of the top keywords are 13058  and 11551 (please see attachment). These values seem very high to me. Am I missing something?

Thanks in advance, and thank you Laurence for this tool and all the resources you made freely available!

Eugenia

[Laurence Anthony]

Hi Eugenia,

Such high values seem completely normal to me. But, the files you sent are not ordered by keyness values, with 0 scores appearing throughout.

Laurence.

###############################################################
Laurence ANTHONY, Ph.D.
Professor of Applied Linguistics
Faculty of Science and Engineering
Waseda University
3-4-1 Okubo, Shinjuku-ku, Tokyo 169-8555, Japan
E-mail: antho...@gmail.com
WWW: http://www.laurenceanthony.net/
###############################################################

--
You received this message because you are subscribed to the Google Groups "AntConc-Discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to antconc+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/antconc/e187b73a-ec8c-4a60-862f-4984aefd8b87n%40googlegroups.com.

[Eugenia Diegoli]

Dear Laurence, 

thank you for your prompt reply! That's reassuring :)

Perhaps the results are not displayed correctly in the txt file? They should be ordered by keyness values (see screenshot).

Thanks again for your help!

Eugenia

To view this discussion visit https://groups.google.com/d/msgid/antconc/CAL6Fgv3a4TUVAggr0Tr1oGniHwYCYVsFmX6GzqX_JuVhyLPv2A%40mail.gmail.com.