Research:Cartograph: Spatial Visualization of Wikipedia Concepts

From Meta, a Wikimedia project coordination wiki
Created
05:07, 10 October 2016 (UTC)
Collaborators
Bret Jackson
Duration:  2016-06 – 2017-06
visualization search gender
This page documents a completed research project.


Screenshot of the Cartograph Spatial Mapping System

The Cartograph visualization system harnesses the vast amount of world knowledge encoded within Wikipedia to create maps of any data. The first map visualizes gender focus in Wikipedia articles. Would you test it out and let me know your thoughts?


Methods[edit]

Cartograph extends previous systems that visualize non-spatial data using geographic approaches. While these systems required data with an existing semantic structure, Cartograph unlocks spatial visualization for any data by enhancing it with semantic information extracted from Wikipedia. Cartograph's map embeddings use neural networks trained on Wikipedia article content and user navigation behavior. Using these embeddings, the system can reveal patterns between points that are unrelated in the original data sets, but are related in meaning and therefore embedded close together on the map.

Timeline[edit]

  • Exploratory user study: Oct 2016
  • Iterative improvement: Oct 2016
  • Public Launch: Nov 2016

Policy, Ethics and Human Subjects Research[edit]

This project has been reviewed and approved by the Institutional Review Board at Macalester College.

Results[edit]

Shilad Sen, Anja Beth Swoap, Qisheng Li, Brooke Boatman, Ilse Dippenaar, Rebecca Gold, Monica Ngo, Sarah Pujol, Bret Jackson, and Brent Hecht (2017) "Cartograph: Unlocking Spatial Visualization Through Semantic Enhancement" in the 22nd International Conference on Intelligent User Interface. doi:10.1145/3025171.3025233.

References[edit]