While this application should still work it's not updated since many years and I think this data can be much better handled using projects such as Wiktextract
This repository is archived to reduce noise
This application exposes a few WordNet functions with a simple REST interface. In particular, can be used from a browser or as a text annotator.
The access to the WordNet database is done using extJWNL
Run the service by cloning it and use mvn compile && mvn exec:java or using Docker
docker run -p 5679:5679 jacopofar/wordnet-as-a-service
Retrieve hypernyms of a word:
curl http://localhost:5679/hypernym/1/fork
[
{
"POS": "noun",
"word": "cutlery"
},
{
"POS": "noun",
"word": "eating utensil"
},
{
"POS": "noun",
"word": "division"
},
{
"POS": "verb",
"word": "lift"
},
...
hyponyms
curl http://localhost:5679/hyponym/1/tool
[
{
"POS": "noun",
"word": "abrader"
},
{
"POS": "noun",
"word": "abradant"
},
{
"POS": "noun",
"word": "bender"
},
...
holonyms
curl http://localhost:5679/holonym/1/France
[
{
"POS": "noun",
"word": "Europe"
},
{
"POS": "noun",
"word": "European Union"
},
{
"POS": "noun",
"word": "EU"
},
...
synonyms, substance holonyms, meronyms and others
Just change the first of the URL with one of: holonym, entailment, substance_meronym, hyponym, antonym, synonym, substance_holonym, meronym, cause or hypernym.
The number in the URL is the senses to be considered when retrieving the word synsets. Increasing it will lead to more results but often seemingly "wrong" ones.
word definitions
Use
http://localhost:5679/definition/can
to get the glosses:
[
{
"POS": "noun",
"gloss": "airtight sealed metal container for food or drink or paint etc.",
"other terms": "[can, tin can, tin]"
},
{
"POS": "noun",
"gloss": "the quantity contained in a can",
"other terms": "[can, canful]"
},
{
...
As an annotator
The server can also produce a list of annotations for a given text and pattern:
curl -X POST -H "Content-Type: application/json" '{"parameter":"pos=n,w=cat","text":"a feline is jumping right here"}' "http://localhost:5679/hypernyms_tagger/1"
will return:
{
"annotations": [
{
"span_start": 2,
"span_end": 8
}
]
}
Generate a sample You can use the same format of the annotator to generate a sample of something matching the word type:
curl -X POST -H "Content-Type: application/json" -H "Cache-Control: no-cache" -H "Postman-Token: 41610ffd-f6e9-1573-850d-86b7bdcb29af" -d '{"parameter":"pos=n,w=bike"}' "http://localhost:5679/sample/hyponym/4"
will return, for example, velocipede.
This format is suitable to be used as an HTTP annotator/generator for Fleximatcher
The server uses Evo inflector to educatedly guess English words plurals and match them (so, matching the hyponyms of "animal" will get "cats" as well as "cat")