LODsyndesisIE: Entity Extraction from Text and Enrichment Using Hundreds of Linked Datasets

We shall demonstrate LODsyndesis IE, which is a research prototype that offers Entity Extraction from text and Entity Enrichment for the extracted entities, using several Linked Datasets. LODsyndesis IE exploits widely used Named Entity Extraction and Disambiguation tools (i.e., DBpedia Spotlight, WAT and Stanford CoreNLP) for identifying the entities of a given text, and enriches each identified entity with hyperlinks to LODsyndesis, which offers various services for millions of entities by leveraging hundreds of Linked Datasets. LODsyndesis IE brings several benefits to the entity extraction task: the user can a) annotate the entities of a given text by selecting different entity recognition tools, b) retrieve all the URIs and facts of each recognized entity from multiple datasets, and c) discover the K most relevant datasets (e.g., datasets containing the most facts) for each entity. The demo is available at https://demos.isl.ics.forth.gr/LODsyndesisIE/.