Web Track
Investigate retrieval behavior on the web
- two tasks
- ad hoc: traditional ad hoc retrieval task
- homepage finding: known-item task to find entry page of site described in topic
- document set
- WT10g collection used in previous years
- 10 GB sample of web pages constructed by ANU/CSIRO from a 1997 spidering obtained from the Internet Archive
- naturally defined subcollections
- some content-heavy pages
- good closed set of links