Each of this scripts takes a text as input and finds the most important entities using dbpedia-spotlight (language: english, confidence: 0.5, support: 0). After that, for every entity, usings its dbpedia page, an abstract is inserting with important informations about the entity. Finally a json file saved in folder "json_files" containing the input text and other information about the annotated entities, including the abstracts.
The node packages which used are:
-
dbpedia-spotlight -- https://github.com/dbpedia-spotlight/DBpediaSpotlight.js/tree/master
-
cheerio -- https://github.com/cheeriojs/cheerio
-
request -- https://github.com/request/request