sebastiano-g / nl2sparql

Natural Language to Sparql Query Language

NL2SPARQL

Natural Language to Sparql Query Language

Competency questions analysis available at this link

The SequenceExtract.py script attempts to isolate the predicates and entities that make up questions, while ExtendedSequenceExtract.py provides meaningful grammatical and syntactic analysis.

Scripts organization

Let's consider the following question "What are the topics of a song?". This question will processed to produce the following outputs:

File	Output
SequenceExtraction.py	{'question': 'Which', 'predicate1': 'are', 'entity1': 'topics', 'entity2': 'song'}
ExtendedSequenceExtract.py	['Which', 'TO-BE-VERB', 'NOUN', 'of', 'NOUN']
LodeExtractor.py	example notebook
ToSPARQL.py	SELECT ?topics WHERE { ?topics <https://songstopoems.github...owl#appears_in> ?song . ?topics hasType <htt://songstopoems.github...owl#Topic> . ?song hasType <https://songstopoems.github...owl#Song> . }