How does Quertle Work?
• An extensive ontology – Consisting of dictionaries, thesauri, hierarchical and non-hierarchical relationships, these manually-curated databases are based upon information extracted from a wide variety of sources. The ontology also contains verbs (e.g., so we know “activate” and “induce” are related) and the collections of related entities (Power Terms). • Entity identification engine – Using the ontology, the entity identification engine recognizes and tags known objects. Recognition of gene and protein names is validated by the content of the surrounding text. • A versatile linguistic engine – This engine uses biomedical-specific natural language processing (NLP) for full-sentence parsing. The processing uses a syntactic parser to find possible sentence structures followed by conversion to a semantic tree. The primary result of that process is the identification of each subject-verb-object triplet, with each triplet normalized to active voice (for instance, “apoptosis is regulated b
