sanskrit-lexicon / hwnorm1

Headword normalization for Cologne dictionaries
0 stars 0 forks source link

ngrams for normalized spellings #11

Open funderburkjim opened 6 years ago

funderburkjim commented 6 years ago

The ngram directory contains 2 and 3 grams computed from the normalized spellings of hwnorm1c. The readme of the ngram directory provides details.

This was done now in order to replace the ngram-checking of simple-search with ngrams drawn from all Cologne Sanskrit dictionaries, rather than just the Monier-Williams dictionary.

gasyoun commented 6 years ago

ngrams drawn from all Cologne Sanskrit dictionaries, rather than just the Monier-Williams dictionary

Now I get the difference as well.