Combining error-correcting output codes and model-refinement for text categorization

Authors:
Songbo Tan;Yuefen Wang
Affiliations:
Institute of Computing Technology, Beijing, China;Chinese Academy of Geological Sciences, Beijing, China
Venue:
SIGIR '07 Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval
Year:
2007

Citing 5
Cited 1

Boosting to correct inductive bias in text classification

Proceedings of the eleventh international conference on Information and knowledge management
Combining Labeled and Unlabeled Data for MultiClass Text Categorization

ICML '02 Proceedings of the Nineteenth International Conference on Machine Learning
A Comparative Study on Feature Selection in Text Categorization

ICML '97 Proceedings of the Fourteenth International Conference on Machine Learning
Using Error-Correcting Codes for Text Classification

ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
A novel refinement approach for text categorization

Proceedings of the 14th ACM international conference on Information and knowledge management

Web page classification: Features and algorithms

ACM Computing Surveys (CSUR)

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this work, we explore the use of error-correcting output codes (ECOC) to enhance the performance of centroid text classifier. The framework is to decompose one multi-class problem into multiple binary problems and then learn the individual binary classification problems by centroid classifier. However, this kind of decomposition incurs considerable bias for centroid classifier, which results in noticeable degradation of performance. To address this issue, we use Model-Refinement to adjust this so-called bias.