DataQuest - Entity Linking

Identify Wikidata entities from web pages and annotate entity mentions directly in the text.

DataQuest is a browser extension that connects web browsing with Wikidata through two independent features: webpage entity identification and inline text annotations. - Identify the main entity of a webpage DataQuest uses webpage URLs and Wikidata mappings to identify the entity associated with the current website when a suitable match is available. Explore information about the identified entity, navigate to related websites, and discover similar entities directly from the extension popup. - Discover entities in webpage text DataQuest also provides text-based Entity Linking. With a single action, users can process webpage paragraphs to identify mentions of people, places, organizations, and other entities. Recognized mentions are annotated directly in the webpage. Hover over an annotation to explore the associated Wikidata entity, including its name, description, identifier, and available links. - Choose how text is processed Select ReFinED, OpenTapioca, or Combined, which uses both entity linking systems. Adjustable batch sizes allow users to configure how longer pages are processed. - Independent features Main-entity identification and text annotations operate independently. You can explore a webpage's main entity without annotating its text, or process annotations even when no main entity is found. - Privacy and user control Text annotation is initiated by the user. Before the first processing request, DataQuest explains that webpage paragraph text is transmitted over HTTPS to a remote entity-linking service hosted on university-operated infrastructure. DataQuest extends the original browser extension developed by Diego Pizarro. The entity linking and inline annotation functionality was developed as part of an academic project at the University of Chile.