Abstract
The ability to maintain awareness within a field of research has been a hallmark of scientific expertise for centuries. As diverse scientific information becomes available through various Internet sources, not merely conference proceedings and journals, maintaining scientific awareness becomes a significant challenge. One challenge is how to discover sources that may be of interest. A second challenge lies in finding significant information over many sources. Scanning through hundreds of posts, feeds, and articles a day is very time consuming and error prone. Our approach uses a set of author published papers as seed documents to recommend documents of interest across various Internet sources. This enables 1) discovery of new sources that may be of interest, and 2) refine the information within a source to only the most relevant.
