Amy Langville, College of Charleston


Nonnegative Matrices in Ranking and Rating


Abstract: There are many applications that rely heavily on ranked lists. For instance, search engines present a ranked list of results to users. Voters rank political candidates in an election. Companies like Netflix produce lists of the highest ranked movies. Algorithms from numerical linear algebra are typically used to create these ranked lists. Different algorithms produce different ranked lists. In addition, they also create rating lists. Related to ranking and rating are the mathematical problems of the distance between ranked lists and rank aggregation, which aims to merge the results from several lists into one aggregated list that contains the best qualities from the various input lists.

Address: Department of Mathematics, College of Charleston, Charleston SC 29424. Go to Professor Langville's website.