Stanford InfoLab Publication Server

Spectral Learning

Kamvar, Kamvar and Sepandar, Sepandar and Klein, Klein and Dan, Dan and Manning, Manning and Christopher, Christopher (2003) Spectral Learning. Technical Report. Stanford InfoLab. (Publication Note: International Joint Conference of Artificial Intelligence)




We present a simple, easily implemented spectral learning algorithm that applies equally whether we have no supervisory information, pairwise link constraints, or labeled examples. In the unsupervised case, it performs consistently with other spectral clustering algorithms. In the supervised case, our approach achieves high accuracy on the categorization of thousands of documents given only a few dozen labeled training documents for the 20 Newsgroups data set. Further more, its classification accuracy increases with the addition of unlabeled documents, demonstrating effective use of unlabeled data.

Item Type:Techreport (Technical Report)
Uncontrolled Keywords:machine learning, spectral methods, clustering, classification, constrained clustering
Related URLs:Project Homepage
ID Code:587
Deposited By:Import Account
Deposited On:19 Apr 2003 17:00
Last Modified:24 Dec 2008 10:12

Download statistics

Repository Staff Only: item control page