Application for knowledge graph enrichment through linking with DBpedia.
pip install requirements.txt
or
python -m pip install requirements.txt
For enriching the ontology, first specify all the properties in the options.json
file:
Name | Description | Example |
---|---|---|
input_file | Path to the ontology in .ttl fromat | "ttl/sto.ttl" |
whitelist | List of predicates to fetch from DBpedia while enriching | [ "<http://purl.org/dc/terms/subject>" ] |
blacklist | List of predicates to exclude from enrichment process | [ "http://purl.org/voc/vrank#hasRank" ] |
prefixes | List of prefixes to assign to the resulting ontology | [ { "prfx": "deo", "uri": "http://purl.org/spar/deo/" } ] |
Next, execute enrich.py
with options.json
as a parameter:
python enrich.py options.json
Using a Turtle format of the Standard Ontology (STO) as an input without specifying whitelist
and blacklist
in the options.json
, i.e., fetching all reachable data from DBpedia, the following results can be achieved:
# of DBpedia links | # of added triples |
---|---|
78 | 8268 |
You can check out the input and the resulting file here.
MIT License. Copyright (c) 2018 Maxim Maltsev.