Minerva II: A Novel Entity Discovery Tool

Entity discovery is a long-lasting interest in governments, enterprises, and the research community. It is a complex task that requires retrieving, extracting, linking, and displaying entities. Algorithms to support entity discovery have been proposed across disciplines including Information Retrieval (IR), Information Extraction (IE), Natural Language Processing (NLP), and Data Mining (DM). However, there is little study on User Interface (UI) for supporting effective entity discovery. This paper presents Minerva II, a novel entity discovery tool, to tackle this challenge. In the paper, we illustrate the UI design and how it effectively supports the typical work flow when a user performs entity discovery. We also describe a new visualization algorithm for entity networks. Our user study shows that Minerva II is able to greatly increase users' efficiency.