The IRMA Community
Newsletters
Research IRM
Click a keyword to search titles using our InfoSci-OnDemand powered search:
|
Using Boosting in a Multilabel Text Categorization Problem
Abstract
The issue of learning to predict the various specific subjects or categories to which individual pieces of text belong is central in automatic classification. The Boosting scheme has been extensively tested and show an interesting record of success. In this paper we examine its performance in a well-known database of Medline research documents known as cystic fibrosis. This database is highly versatile for systematic experimentation. We empirically demonstrate this approach, discuss its overall usefulness and scope, and provide a detailed roadmap of further research in the area.
|
|