Cranking: Combining Rankings Using Conditional Probability Models on Permutations

Guy Lebanon


  A new approach to ensemble learning is introduced that takes ranking rather than classification as fundamental, leading to models on the symmetric group and its cosets. The approach uses a generalization of the Mallows model on permutations to combine multiple input rankings. Applications include the task of combining the output of multiple search engines and multiclass or multilabel classification, where a set of input classifiers is viewed as generating a ranking of class labels. Experiments for both types of applications are presented.

Link to the paper.

This is joint work with John Lafferty.

Back to the Main Page

Charles Rosenberg
Last modified: Mon May 20 13:32:26 EDT 2002