Natural language query formalization to SPARQL for querying knowledge bases using Rasa

Abstract The idea of data to be semantically linked and the subsequent usage this with modern computer applications has been one most important aspects Web 3.0. However, actualization aspect challenging due difficulties associated building knowledge bases using formal languages query them. In regard, SPARQL, a recursive acronym for standard language protocol Linked Open Data Resource Description Framework databases, is popular querying language. Nonetheless, writing SPARQL queries known difficult, even experts. Natural formalization, which involves parsing natural their equivalents, an essential step in overcoming steep learning curve. Recent work field seen artificial intelligence (AI) techniques modelling adequate accuracy. This paper discusses design creating closed domain ontology, then used by AI-powered chat-bot that incorporates formalization Rasa entity extraction after intent recognition. A precision–recall analysis performed in-built tools conjunction our own testing parameters, it found system achieves precision 0.78, recall 0.79 F1-score 0.79, are better than current state art.