ETL ensembles for chunking, NER and SRL

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Authors

We present a new ensemble method that uses Entropy Guided Transformation Learning (ETL) as the base learner. The proposed approach, ETL Committee, combines the main ideas of Bagging and Random Subspaces. We also propose a strategy to include redundancy in transformation-based models. To evaluate the effectiveness of the ensemble method, we apply it to three Natural Language Processing tasks: Text Chunking, Named Entity Recognition and Semantic Role Labeling. Our experimental findings indicate that ETL Committee significantly outperforms single ETL models, achieving state-of-the-art competitive results. Some positive characteristics of the proposed ensemble strategy areworth to mention. First, it improves the ETL effectiveness without any additional human effort. Second, it is particularly useful when dealing with very complex tasks that use large feature sets. And finally, the resulting training and classification processes are very easy to parallelize.

Original languageEnglish
Title of host publicationComputational Linguistics and Intelligent Text Processing : 11th International Conference, CICLing 2010, Iaşi, Romania, March 21-27, 2010. Proceedings
EditorsAlexander Gelbukh
Number of pages13
Place of PublicationBerlin
PublisherSpringer
Publication date2010
Pages100-112
ISBN (print)3-642-12115-2, 978-3-642-12115-9
ISBN (electronic)978-3-642-12116-6
DOIs
Publication statusPublished - 2010
Externally publishedYes
Event11th International Conference on Computational Linguistics and Intelligent Text Processing - CICLing 2010 - Universität Alexandru Ioan Cuza, Iasi, Romania
Duration: 21.03.201027.03.2010
Conference number: 11
http://www.cicling.org/2010/

    Research areas

  • Ensemble methods, Entropy guided transformation learning, Named entity recognition, Semantic role labeling, Text chunking
  • Informatics
  • Business informatics