About Us Our Work Employment News & Events
MITRE Remote Access for MITRE Staff and Partners Site Map
Our Work

Follow Us:

Visit MITRE on Facebook
Visit MITRE on Twitter
Visit MITRE on Linkedin
Visit MITRE on YouTube
View MITRE's RSS Feeds
View MITRE's Mobile Apps
Home > Our Work > Technical Papers >

Coaxing Confidences From an Old Friend: Probabilistic Classifications From Transformation Rule List

August 2000

John C. Henderson, The MITRE Corporation
Radu Florian, Johns Hopkins University
Grace Ngai, Johns Hopkins University

ABSTRACT

Transformation-based learning has been successfully employed to solve many natural language processing problems. It has many positive features, but one drawback is that it does not provide estimates of class membership probabilities.

In this paper, we present a novel method for obtaining class membership probabilities from a transformation-based rule list classifier. Three experiments are presented which measure the modeling accuracy and cross-entropy of the probabilistic classifier on unseen data and the degree to which the output probabilities from the classifier can be used to estimate confidences in its classification decisions.

The results of these experiments show that, for the task of text chunking*, the estimates produced by this technique are more informative than those generated by a state-of-the-art decision tree.

*All the experiments are performed on text chunking. The technique presented is general-purpose, however, and can be applied to many tasks for which transformation-based learning performs well, without changing the internals of the learner.

View/Download Document

Additional Search Keywords

N/A

Page last updated: August 10, 2000   |   Top of page

Homeland Security Center Center for Enterprise Modernization Command, Control, Communications and Intelligence Center Center for Advanced Aviation System Development

 
 
 

Solutions That Make a Difference.®
Copyright © 1997-2013, The MITRE Corporation. All rights reserved.
MITRE is a registered trademark of The MITRE Corporation.
Material on this site may be copied and distributed with permission only.

IDG's Computerworld Names MITRE a "Best Place to Work in IT" for Eighth Straight Year The Boston Globe Ranks MITRE Number 6 Top Place to Work Fast Company Names MITRE One of the "World's 50 Most Innovative Companies"
 

Privacy Policy | Contact Us