A data visualization framework to leverage text and knowledge graphs

The goal of this work will be to explore different ways to visualize and interact with knowledge extracted automatically from very large heterogeneous document collections. This extracted knowledge will be in the form of a multi-attribute graph of extracted entities and relationships between them. These relationships will be associated with both temporal and spatial information. The work conducted will focus in identifying the best ways to visually represent and interact with content from two application domains — medicine and journalism — both including thousands of entities and relationships. By leveraging these very distinct domains, we aim to provide a unified framework and initial prototype to navigate large multivariate knowledge graphs that is potentially applicable across multiple domains.

Faculty Supervisor:

Fanny Chevalier;Michael Brudno

Student:

Partner:

Université Paris Saclay

Discipline:

Computer science

Sector:

New and Digital Media; Information and Communications Technology; Health and Related Sciences & Technology

University:

University of Toronto

Program: