This repository contains a collection of scripts for
- collecting Wikipedia + Flickr/Twitter/YouTube accounts and the edits and posts of those users
- detecting named entities contained within those text-based utterances on these social-Web sites
- modeling topics and user interest with respect to the Wikipedia categorization system
- applying user interest model to rank candidates of an ambiguous entity detected in a user's post
Contact: Elizabeth Murnane [email protected]