Kevin Heinrich and
Michael W. Berry,
University of Tennessee;
Ramin Homayouni,
Universtiy of Tennessee Health Science Center
10:30 AM - 12:30 PM
Room: Magnolia - 3rd Floor
This symposium covers three popular algorithms in information retrieval. All three algorithms rely on eigenvector methods. The famous PageRank algorithm of Google, which computes the dominant eigenvector of an enormous sparse Markov chain, ranks webpages according to values in this eigenvector. A related algorithm, HITS, ranks webpages according to values in two dominant eigenvectors. And, the well-established algorithm, LSI, uses the singular value decomposition of a large sparse nonnegative matrix to answer user queries and cluster documents. This symposium starts with an overview of these methods, followed by some novel applications of and modifications to the three basic algorithms.
Organizer:
Carl Meyer
North Carolina State University
Amy Langville
College of Charleston
Kevin Heinrich and
Michael W. Berry,
University of Tennessee;
Ramin Homayouni,
Universtiy of Tennessee Health Science Center