java - Stanford NLP/NER - how to use extractors and classifiers? -

java - Stanford NLP/NER - how to use extractors and classifiers? -

the demo file comes stanford's named entity recognition bundle shows examples of classifying sentences. outputs options strings such as:

"i eat apples michael , hashemite kingdom of jordan on mondays"

=>

"i eat apples <person>michael</person> , <person>jordan</person> on <date>mondays</date>"

or map of classifications per word.

is there method returns map of classification list of entities?

eg:

{ person : ["michael", "peter"] date : ["mondays"] }

here minimal bash script postprocessing :

echo "i eat apples <person>michael</person> , <person>jordan</person> on <date>mondays</date>" | grep -eo '<([^>/]*)>[^<]*' | awk -f '>' '{nes[$1]=nes[$1]","$2;} end {for(t in nes) print tolower(t)":{"nes[t]"}";}' | sed 's/:{,/:{/' | tr -d '<'

using you'll obtain lists illustration :

date:{mondays} person:{michael,jordan}

you'll want handle multiple entries ("uniq" or count them?).

java stanford-nlp named-entity-recognition named-entity-extraction