We present results on using stacking to ensemble multiple systems for the Knowledge Base Population English Slot Filling (KBP-ESF) task. In addition to using the output and confidence of each system as input to the stacked classifier, we also use features capturing how well the systems agree about the provenance of the information they extract. We demonstrate that our stacking approach outperforms the best system from the 2014 KBP-ESF competition as well as alternative ensembling methods employed in the 2014 KBP Slot Filler Validation task and several other ensembling baselines. Additionally, we demonstrate that including provenance information further increases the performance of stacking.

PDF

In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics (ACL-15), pp. 177-187, Beijing, China, July 2015.

@inproceedings{viswanathan:acl15, title={Stacked Ensembles of Information Extractors for Knowledge-Base Population}, author={Vidhoon Viswanathan and Nazneen Fatema Rajani and Yinon Bentor and Raymond J. Mooney}, booktitle={Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics (ACL-15)}, month={July}, address={Beijing, China}, pages={177-187}, url="http://www.cs.utexas.edu/users/ai-labpub-view.php?PubID=127515", year={2015} }

Slides (PPT)