We demonstrate the merits of using inter-document similarities for federated search. Specifically, we study a resultsmerging method that utilizes information induced from cluster...
Most existing reranking approaches to image search focus solely on mining "visual" cues within the initial search results. However, the visual information cannot always ...
The Internet is widely valued for distributing control over information to a lateral network of individuals, but it is not clear how these networks can most effectively organize t...
Cristen Torrey, Moira Burke, Matthew L. Lee, Anind...
Large quantities of documents in the Internet and digital libraries are simply scanned and archived in image format, many of which are packed in PDF files. The word search tool pr...
Fully automatic methods that extract lists of objects from the Web have been studied extensively. Record extraction, the first step of this object extraction process, identifies...