nlp - What other inputs are there to Word Sense Disambiguation task? -
in natural language processing
(nlp), word sense disambiguation
(wsd) task computationally determines meaning(s) or sense(s) or concept(s) of polysemous word given sentence word appears in. example:
- "some stupid enough rob central bank*."*
- "the river bank is full of stones"
do know on wsd performed in paragraph or document level?
other disambiguate senses/meaning context words in 1 sentence, what other input introduce perform wsd
task? (i've seen wsd images before, http://acl.ldc.upenn.edu/w/w03/w03-0601.pdf)
when start talking paragraphs or documents, times rather word senses interested in topics. use collection of word senses assigned words in paragraph figure out topic, problems connected, although it's better think in terms of topics rather senses (which associated words or perhaps phrases).
good luck, ted
Comments
Post a Comment