Organizing and searching the world wide web of facts -- step two: harnessing the wisdom of the crowds Export

@inproceedings{citeulike:1587018, abstract = {As part of a large effort to acquire large repositories of facts from unstructured text on the Web, a seed-based framework for textual information extraction allows for weakly supervised extraction of class attributes (e.g., side effects and generic equivalent for drugs) from anonymized query logs. The extraction is guided by a small set of seed attributes, without any need for handcrafted extraction patterns or further domain-specific knowledge. The attributes of classes pertaining to various domains of interest to Web search users have accuracy levels significantly exceeding current state of the art. Inherently noisy search queries are shown to be a highly valuable, albeit unexplored, resource for Web-based information extraction, in particular for the task of class attribute extraction.}, address = {New York, NY, USA}, author = {Pa\c{s}ca, Marius}, booktitle = {WWW '07: Proceedings of the 16th international conference on World Wide Web}, citeulike-article-id = {1587018}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1242587}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/1242572.1242587}, doi = {10.1145/1242572.1242587}, isbn = {978-1-59593-654-7}, keywords = {crowds, pish, teeth, wisdom, wish, www}, location = {Banff, Alberta, Canada}, pages = {101--110}, posted-at = {2009-09-18 21:34:03}, priority = {2}, publisher = {ACM}, title = {Organizing and searching the world wide web of facts -- step two: harnessing the wisdom of the crowds}, url = {http://dx.doi.org/10.1145/1242572.1242587}, year = {2007} }

TY - CONF ID - citeulike:1587018 L3 - citeulike-article-id:1587018 N2 - As part of a large effort to acquire large repositories of facts from unstructured text on the Web, a seed-based framework for textual information extraction allows for weakly supervised extraction of class attributes (e.g., side effects and generic equivalent for drugs) from anonymized query logs. The extraction is guided by a small set of seed attributes, without any need for handcrafted extraction patterns or further domain-specific knowledge. The attributes of classes pertaining to various domains of interest to Web search users have accuracy levels significantly exceeding current state of the art. Inherently noisy search queries are shown to be a highly valuable, albeit unexplored, resource for Web-based information extraction, in particular for the task of class attribute extraction. CY - Banff, Alberta, Canada SN - 978-1-59593-654-7 AD - New York, NY, USA EP - 110 TI - Organizing and searching the world wide web of facts -- step two: harnessing the wisdom of the crowds PB - ACM T2 - WWW '07: Proceedings of the 16th international conference on World Wide Web SP - 101 KW - crowds KW - pish KW - teeth KW - wisdom KW - wish KW - www AU - Paşca, Marius PY - 2007/// UR - http://dx.doi.org/10.1145/1242572.1242587 ER -