Speaker: Alexandre Klementiev
 
Affiliation: UIUC (CS)

Title: Unsupervised Rank Aggregation with Distance-Based Models
 
Abstract:
The need to meaningfully combine sets of rankings often comes up when one deals with ranked 
data.  Although a number of heuristic and supervised learning approaches to rank aggregation 
exist, they require domain knowledge or supervised ranked data, both of which are expensive to 
acquire.  In order to address these limitations, we propose a mathematical and algorithmic 
framework for learning to aggregate (partial) rankings without supervision.  In many practical 
applications, constituent rankers are likely to specialize in some input domains, while 
performing poorly in others.  Therefore, we extend our framework to take such latent domain 
information into account.

  

We instantiate the framework for the cases of combining permutations and combining top-k 
lists, and experimentally demonstrate the effectiveness of the proposed formalism in both 
scenarios.  We also show that that the extended framework can take advantage of the latent 
domain information to learn to aggregate the votes of rankers with domain-specific expertise.
 
Bio:
Alexandre Klementiev is a Ph.D. student at UIUC Computer Science Department working with 
Prof. Dan Roth.  His current research interests are on the intersection of Machine Learning 
and Natural Language Processing.
 
back