Record Details

Discriminative methods for multi-labeled classification

DSpace at IIT Bombay

View Archive Info
 
 
Field Value
 
Title Discriminative methods for multi-labeled classification
 
Creator GODBOLE, S
SARAWAGI, S
 
Description In this paper we present methods of enhancing existing discriminative classifiers for multi-labeled predictions. Discriminative methods like support vector machines perform very well for uni-labeled text classification tasks. Multi-labeled classification is a harder task subject to relatively less attention. In the multi-labeled setting, classes are often related to each other or part of a is-a hierarchy. We present a new technique for combining text features and features indicating relationships between classes, which can be used with any discriminative algorithm. We also present two enhancements to the margin of SVMs for building better models in the presence of overlapping classes. We present results of experiments on real world text benchmark datasets. Our new methods beat accuracy of existing methods with statistically significant improvements.
 
Publisher SPRINGER-VERLAG BERLIN
 
Date 2011-10-23T15:42:57Z
2011-12-15T09:11:13Z
2011-10-23T15:42:57Z
2011-12-15T09:11:13Z
2004
 
Type Article; Proceedings Paper
 
Identifier ADVANCES IN KNOWLEDGE DISCOVERY AND DATA MINING, PROCEEDINGS,3056,22-30
3-540-22064-X
0302-9743
http://dspace.library.iitb.ac.in/xmlui/handle/10054/15166
http://hdl.handle.net/100/1929
 
Source 8th Pacific/Asia Conference on Advances in Knowledge Discovery and Data Mining,Sydney, AUSTRALIA,MAY 26-28, 2004
 
Language English