I was in CISTI in Ottawa yesterday and saw an interesting and regrettably brief presentation from Glen Newton about the work of their research group They are doing some nice things to layer useful functionality across large data sets (e.g. clustering, mapping citation patterns, recommendation, ...).
Some examples of their work are visible through CISTI Lab:
CISTI Lab is an experimental site for demonstrating and evaluating prototype software and services developed by CISTI staff and research partners. [NRC-CISTI, Welcome to CISTI Lab]
Projects are linked from a Wiki. They include Ungava:
Ungava - explores ways of navigating full-text search and the visualization of search results for articles in the NRC Research Press collection (demo) and the Colorado State University Libraries catalogue (demo). Includes a drill cloud implementation. [Main Page - CISTI-ICIST LAB WIKI]
The public test collections are smallish. Ungava is built with Lucene and incorporates some work from the Simile project at MIT, including the exhibit and timeline. It implements Coins. It also uses what it calls drill clouds:
Ungava extends tag clouds to make them a useful tool for search refinement. That is, to use a tag cloud to refine an existing query by adding new elements to the query through interactions with the cloud. As this results in a kind of drill-down search behaviour, these new clouds have been named drill clouds. [Drill Clouds - CISTI-ICIST LAB WIKI]