CACM logo

This Article

Research highlights

Predicting Structured Objects with Support Vector Machines

[article image]
Some red and blue points in the hyperplane image are marked with yellow dots. In support vector machines, a predictive model is not parameterized in terms of the weights assigned to features but in terms of weights associated with each case. Credit: Columbia Univ. Dept. of Statistics

Thorsten Joachims, Thomas Hofmann, Yisong Yue, Chun-Nam Yu

November 1, 2009

Machine Learning today offers a broad repertoire of methods for classification and regression. But what if we need to predict complex objects like trees, orderings, or alignments? Such problems arise naturally in natural language processing, search engines, and bioinformatics. The following explores a generalization of Support Vector Machines (SVMs) for such complex prediction problems.

Read the Full Article:
Abstract  |  Full Text (HTML)  |  Full Text (PDF)  |  In the Digital Edition  |  In the Digital Library

Tools For Readers

Bookmark and Share
Default Font Size Large Font Size X-Large Font Size Text Size
Author profiles and more in the Digital Library

Related ACM Resources

Conferences:

Books:

Courses:

  • Project Management for Technical Teams - In this course, you will identify methods of effectively managing small- to medium-sized projects and achieving their stated objectives. (Duration: 113 minutes)
More About ACM Resources

In The Digital Library

About Communications | Join ACM External Link | Renew External Link | Subscribe External Link | Sign In | For Authors | For Advertisers External Link | Privacy | Site Map | Help | Contact Us

Copyright © 2009 by the ACM. All rights reserved.

Back to Top