An example-based mapping method for text categorization and retrieval
ACM Transactions on Information Systems (TOIS)
SIGIR '94 Proceedings of the 17th annual international ACM SIGIR conference on Research and development in information retrieval
The nature of statistical learning theory
The nature of statistical learning theory
Method combination for document filtering
SIGIR '96 Proceedings of the 19th annual international ACM SIGIR conference on Research and development in information retrieval
Combining classifiers in text categorization
SIGIR '96 Proceedings of the 19th annual international ACM SIGIR conference on Research and development in information retrieval
Training algorithms for linear text classifiers
SIGIR '96 Proceedings of the 19th annual international ACM SIGIR conference on Research and development in information retrieval
Inductive learning algorithms and representations for text categorization
Proceedings of the seventh international conference on Information and knowledge management
Using a generalized instance set for automatic text categorization
Proceedings of the 21st annual international ACM SIGIR conference on Research and development in information retrieval
A re-examination of text categorization methods
Proceedings of the 22nd annual international ACM SIGIR conference on Research and development in information retrieval
Text Categorization with Suport Vector Machines: Learning with Many Relevant Features
ECML '98 Proceedings of the 10th European Conference on Machine Learning
Combining Multiple Learning Strategies for Effective Cross Validation
ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
| Hi-index | 0.00 |
We investigate two meta-model approaches for the task of automatic textual document categorization. The first approach is the linear combination approach. Based on the idea of distilling the characteristics of how we estimate the merits of each component algorithm, we propose three different strategies for the linear combination approach. The linear combination approach makes use of limited knowledge in the training document set. To address this limitation, we propose the second meta-model approach, called Meta-learning Using Document Feature characteristics (MUDOF), which employs a meta-learning phase using document feature characteristics. Document feature characteristics, derived from the training document set, capture some inherent properties of a particular category. Extensive experiments have been conducted on a real-world document collection and satisfactory performance is obtained.