Flammie A Pirinen on github pages
an informal survey into meanings of the ‘morphological complexity’ in the context of NLP.
This is a pet peeve of mine and a parallel of What is a low resource language in NLP page I wrote earlier. In practice, I collect citations of scientists who write about morphological complexity in NLP.
Shortly put, morphological complexity is a scale that ranges from nearly no productive morphological processes like in Mandarin Chinese or English to OMG all my sentences are just 1 (or two, or three) complex words of the poly-synthetic languages like Greenlandic, Navajo and what have you. And everything in between this are on the scale somewhere. But 95 % of the language are very very near to English on the scale, and very very far from Greenlandic. Calling languages like French morphologically complex just because it has less than ten more suffixes in one layer than English does is ludicrous. But I digress, this is a collection of more or less accurate usages of ‘morphologically complex’ in published NLP articles.
This is also tongue-in-cheek, not meant to insult any of the authors, in no particular order etc. etc.
We synthesize the contributions of researchers working on parsing Arabic, Basque, French, German, Hebrew, Hindi and Korean
#BenderRule: