statistics - Githubissues

lingpy / morphopy

Python library for handling morphological relations in historical linguistics

GNU General Public License v3.0

0 stars 0 forks source link

Open LinguList opened 5 years ago

LinguList commented 5 years ago

counts, word family size, etc.

Schweikhard commented 5 years ago

Schweikhard commented 5 years ago

Both for statistics and for checking:

[ ] receive an overview of words that share at least one cogID
[ ] receive an overview of words that share at least one crossID
[ ] receive an overview of words that share at least one rootID Those could be sorted by ID. Thereby one would have summaries of the word families (of different levels) in the dataset.

Schweikhard commented 5 years ago

[ ] receive an overview visualization of word families in the data - basically what edictor provides, but with the option to ignore partial cognacy caused by sharing grammatical morphemes.

LinguList commented 5 years ago

Okay, this is what I'll add today to morphopy. A command:

$ morphopy word-families wordlist

will spit out the word families, in some fashion, ignoring grammar shared...