This work addresses relevance feedback as an alternative to keyword-based search engines for sifting through large PDF document collections and extracting the most relevant documents (especially for literature review purposes). Until now, relevance feedback has only been used in content-based image and video retrieval due to the inability to query those media types without keywords. Since PDF journal articles contain many valuable non-keyword features such as structure and formatting information as well as embedded figures, they would benefit from relevance feedback. Stripping a PDF into “full-text” for indexing purposes disregards these important features. We investigate how they can be used to our advantage and look to integrate the wealth of knowledge from relevance feedback text-based information retrieval. We argue for the benefits of placing the burden of relevance judgment on the user rather than the retrieval system and present alternative document views that quickly allow the user to deem relevance.

 Project Sponsors:

 Collaborators:

 Graduate Students:

 Ammon Christiansen and Yuchou Chang

Publications:

1. A.J. Christiansen, D.J. Lee, and Y. Chang, “Finding Relevant PDF Medical Journal Articles by the Caption and Content of Their Figures”, SPIE Medical Imaging, Picture Archive and Communication Systems (PACS) and Imaging Informatics, Vol. 6516-17, San Diego, CA, USA, February 17-22, 2007.

2. A.J. Christiansen and D.J. Lee, “Relevance Feedback Query Refinement for PDF Medical Journal Articles”, Proceedings of The 19th IEEE Symposium on Computer-Based Medical Systems, p. 57-62, Salt Lake City, Utah, June 22-23, 2006.

(Click image to view)