This application generates a Gephi network graph file of persons which are found in pre-specified english input texts. Two persons are in the same network if they are being found in the same document. It reads resources as text files, reads them in to a String, extracts names with Named Entity Extraction and a large list of prenames and generates based on these found names a GEXF file (Graph Exchange XML Format).
##configuration Edit Config.java:
- input text files - texts should be in english and plain text
- export path of graph file – keep .gexf extension in file path
##testing
mvn test
##building
assemble .jar with dependencies with mvn clean compile assembly:single
##extending Since it's based reading a String, more input data types can be added later on (web-fetching, databases, ...)
##example Forbes Top 10 of the richest persons' wikipedia pages (Wikipedia: The world's billionaires):
More examples in examples
folder.