: We describe our participation in the TREC 2007 Enterprise track and detail our language modeling-based approaches. For document search, our focus was on estimating a mixture mode...
Co-training is a semi-supervised technique that allows classifiers to learn with fewer labelled documents by taking advantage of the more abundant unclassified documents. However, ...
In this paper, we introduce the concept of "user policies" and its applications to the browsing of HTML documents. The objective of policies is to specify user preferenc...
This paper describes the development of algorithms for extracting the title and the names of the authors from documents available on the World Wide Web. In this paper we describe ...
Eric G. Berkowitz, Mohamed Reda Elkhadiri, Tim Sah...
A key advantage of hypermedia systems such as the Web is that the user is able to navigate through the information space in a non-linear fashion. He can explore the interlinked do...