Skip Navigation
York U: Redefine the PossibleHOME | Current Students | Faculty & Staff | Research | International
Search »FacultiesLibrariesCampus MapsYork U OrganizationDirectorySite Index
Future Students, Alumni & Visitors
2005 Technical Reports

A Framework for Unsupervised Learning With Multiple Criteria

Bill Andreopoulos, Aijun An and Xiaogang Wang

Technical Report CS-2005-10

York University

May 2005


Unsupervised learning algorithms that have been presented in the literature so far, rarely incorporate more than one criterion for learning. Moreover, the results from learning based on old criteria are not incorporated into the process of learning based on new criteria. This paper presents a framework for unsupervised learning called Doctris "Double Criteria Triple Step". Learning algorithms based on Doctris use two different criteria at two levels of learning. The first level forms a prior for the second level, simulating a Bayesian process. We refer to this process as multi-criteria unsupervised learning. We justify Doctris with the Bayesian theory of classification. We present a metric for evaluating the results. We discuss extending Doctris to use more than two criteria. We describe the BILCOM clustering algorithm as an example of Doctris. Experimental results from clustering mixed data types with BILCOM indicate that it partitions mixed data accurately.

Notice:The work presented in the paper above is covered by pending patents and copyright. Publication of this paper does not grant rights to any intellectual property. All rights reserved.

Download paper in PDF format.

The documents distributed by this server have been provided by the contributing authors as a means to ensure timely dissemination of scholarly and technical work on a noncommercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.