Automatic document metadata extraction using support vector machines Export

@inproceedings{citeulike:1158486, abstract = {Automatic metadata generation provides scalability and usability for digital libraries and their collections. Machine learning methods offer robust and adaptable automatic metadata extraction. We describe a Support Vector Machine classification-based method for metadata extraction from header part of research papers and show that it outperforms other machine learning methods on the same task. The method first classifies each line of the header into one or more of 15 classes. An iterative convergence procedure is then used to improve the line classification by using the predicted class labels of its neighbor lines in the previous round. Further metadata extraction is done by seeking the best chunk boundaries of each line. We found that discovery and use of the structural patterns of the data and domain based word clustering can improve the metadata extraction performance. An appropriate feature normalization also greatly improves the classification performance. Our metadata extraction method was originally designed to improve the metadata extraction quality of the digital libraries Citeseer [17] and EbizSearch[24]. We believe it can be generalized to other digital libraries.}, address = {Washington, DC, USA}, author = {Han, Hui and Giles, Lee C. and Manavoglu, Eren and Zha, Hongyuan and Zhang, Zhenyue and Fox, Edward A.}, booktitle = {JCDL '03: Proceedings of the 3rd ACM/IEEE-CS joint conference on Digital libraries}, citeulike-article-id = {1158486}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=827140.827146}, isbn = {0769519393}, keywords = {automatic\_metadata\_generation, citeseer, support\_vector\_machines}, pages = {37--48}, posted-at = {2007-06-07 18:48:35}, priority = {0}, publisher = {IEEE Computer Society}, title = {Automatic document metadata extraction using support vector machines}, url = {http://portal.acm.org/citation.cfm?id=827140.827146}, year = {2003} }

TY - CONF ID - citeulike:1158486 L3 - citeulike-article-id:1158486 N2 - Automatic metadata generation provides scalability and usability for digital libraries and their collections. Machine learning methods offer robust and adaptable automatic metadata extraction. We describe a Support Vector Machine classification-based method for metadata extraction from header part of research papers and show that it outperforms other machine learning methods on the same task. The method first classifies each line of the header into one or more of 15 classes. An iterative convergence procedure is then used to improve the line classification by using the predicted class labels of its neighbor lines in the previous round. Further metadata extraction is done by seeking the best chunk boundaries of each line. We found that discovery and use of the structural patterns of the data and domain based word clustering can improve the metadata extraction performance. An appropriate feature normalization also greatly improves the classification performance. Our metadata extraction method was originally designed to improve the metadata extraction quality of the digital libraries Citeseer [17] and EbizSearch[24]. We believe it can be generalized to other digital libraries. SN - 0769519393 AD - Washington, DC, USA EP - 48 TI - Automatic document metadata extraction using support vector machines PB - IEEE Computer Society T2 - JCDL '03: Proceedings of the 3rd ACM/IEEE-CS joint conference on Digital libraries SP - 37 KW - automatic_metadata_generation KW - citeseer KW - support_vector_machines AU - Han, Hui AU - Giles, Lee AU - Manavoglu, Eren AU - Zha, Hongyuan AU - Zhang, Zhenyue AU - Fox, Edward PY - 2003/// UR - http://portal.acm.org/citation.cfm?id=827140.827146 ER -