Co-training and Self-training for Word Sense Disambiguation

PDF Version Also Available for Download.

Description

This paper investigates the application of co-training and self-training to word sense disambiguation. Optimal and empirical parameter selection methods for co-training and self-training are investigated, with various degrees of error reduction. A new method that combines co-training with majority voting is introduced, with the effect of smoothing the bootstrapping learning curves, and improving the average performance.

Physical Description

8 p.

Creation Information

Mihalcea, Rada, 1974- May 2004.

Context

This paper is part of the collection entitled: UNT Scholarly Works and was provided by UNT College of Engineering to Digital Library, a digital repository hosted by the UNT Libraries. It has been viewed 129 times . More information about this paper can be viewed below.

Who

People and organizations associated with either the creation of this paper or its content.

Provided By

UNT College of Engineering

The UNT College of Engineering promotes intellectual and scholarly pursuits in the areas of computer science and engineering, preparing innovative leaders in a variety of disciplines. The UNT College of Engineering encourages faculty and students to pursue interdisciplinary research among numerous subjects of study including databases, numerical analysis, game programming, and computer systems architecture.

Contact Us

What

Descriptive information to help identify this paper. Follow the links below to find similar items on the Digital Library.

Degree Information

Description

This paper investigates the application of co-training and self-training to word sense disambiguation. Optimal and empirical parameter selection methods for co-training and self-training are investigated, with various degrees of error reduction. A new method that combines co-training with majority voting is introduced, with the effect of smoothing the bootstrapping learning curves, and improving the average performance.

Physical Description

8 p.

Source

  • Conference on Natural Language Learning (CoNLL), 2004, Boston, Massachusetts, United States

Language

Item Type

Collections

This paper is part of the following collection of related materials.

UNT Scholarly Works

The Scholarly Works Collection is home to materials from the University of North Texas community's research, creative, and scholarly activities and serves as UNT's Open Access Repository. It brings together articles, papers, artwork, music, research data, reports, presentations, and other scholarly and creative products representing the expertise in our university community. Access to some items in this collection may be restricted.

What responsibilities do I have when using this paper?

When

Dates and time periods associated with this paper.

Creation Date

  • May 2004

Added to The UNT Digital Library

  • Jan. 31, 2011, 2:01 p.m.

Description Last Updated

  • June 21, 2013, 10:32 a.m.

Usage Statistics

When was this paper last used?

Yesterday: 0
Past 30 days: 0
Total Uses: 129

Interact With This Paper

Here are some suggestions for what to do next.

Start Reading

PDF Version Also Available for Download.

Citations, Rights, Re-Use

Mihalcea, Rada, 1974-. Co-training and Self-training for Word Sense Disambiguation, paper, May 2004; (digital.library.unt.edu/ark:/67531/metadc30955/: accessed May 27, 2017), University of North Texas Libraries, Digital Library, digital.library.unt.edu; crediting UNT College of Engineering.