Harvesting Relations from the Web – Quantifiying the Impact of Filtering Functions – Export

@article{citeulike:2191982, abstract = {Several bootstrapping-based relation extraction algorithms working on large corpora or on the Web have been presented in the literature. A crucial issue for such algorithms is to avoid the introduction of too much noise into further iterations. Typically, this is achieved by applying appropriate pattern and tuple evaluation measures, henceforth called fil- tering functions, thereby selecting only the most promising patterns and tuples. In this paper, we systematically compare different filtering functions proposed across the literature. Although we also discuss our own implementation of a pattern learning algorithm, the main contribution of the paper is actually the extensive comparison and evaluation of the different filtering functions proposed in the literature with respect to seven datasets. Our results indicate that some of the commonly used filters do not outperform a trivial baseline filter in a statistically significant manner.}, author = {Blohm, S. and Cimiano, P. and Stemle, E.}, booktitle = {Proceedings of the Twenty-Second AAAI Conference on Artificial Intelligence}, citeulike-article-id = {2191982}, citeulike-linkout-0 = {http://www.aifb.uni-karlsruhe.de/WBS/seb/publications/pronto-aaai07.pdf}, keywords = {ie, relationextraction, textmining}, month = {July}, posted-at = {2008-01-03 18:47:59}, priority = {4}, title = {Harvesting Relations from the Web – Quantifiying the Impact of Filtering Functions –}, url = {http://www.aifb.uni-karlsruhe.de/WBS/seb/publications/pronto-aaai07.pdf}, year = {2007} }

TY - JOUR ID - citeulike:2191982 L3 - citeulike-article-id:2191982 N2 - Several bootstrapping-based relation extraction algorithms working on large corpora or on the Web have been presented in the literature. A crucial issue for such algorithms is to avoid the introduction of too much noise into further iterations. Typically, this is achieved by applying appropriate pattern and tuple evaluation measures, henceforth called fil- tering functions, thereby selecting only the most promising patterns and tuples. In this paper, we systematically compare different filtering functions proposed across the literature. Although we also discuss our own implementation of a pattern learning algorithm, the main contribution of the paper is actually the extensive comparison and evaluation of the different filtering functions proposed in the literature with respect to seven datasets. Our results indicate that some of the commonly used filters do not outperform a trivial baseline filter in a statistically significant manner. TI - Harvesting Relations from the Web – Quantifiying the Impact of Filtering Functions – T2 - Proceedings of the Twenty-Second AAAI Conference on Artificial Intelligence KW - ie KW - relationextraction KW - textmining AU - Blohm, S AU - Cimiano, P AU - Stemle, E PY - 2007/07// UR - http://www.aifb.uni-karlsruhe.de/WBS/seb/publications/pronto-aaai07.pdf ER -