Page-level template detection via isotonic smoothing Export

@inproceedings{citeulike:1288655, address = {New York, NY, USA}, author = {Chakrabarti, Deepayan and Kumar, Ravi and Punera, Kunal}, booktitle = {WWW '07: Proceedings of the 16th international conference on World Wide Web}, citeulike-article-id = {1288655}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1242572.1242582}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/1242572.1242582}, citeulike-linkout-2 = {http://www2007.org/paper588.php}, comment = {Simple approach: hash fragment, check frequencies. Approach here: site-level. No training data required. The function that is learned goes from DOM nodes to {template/nontemplate}. Features: placement, bgcolor, aspect ratio, link density, avg. sentence site, etc. Smoothing: all the children of a template are templates. }, doi = {10.1145/1242572.1242582}, isbn = {9781595936547}, keywords = {classification, similarity, text}, pages = {61--70}, posted-at = {2007-05-10 17:38:06}, priority = {0}, publisher = {ACM Press}, title = {Page-level template detection via isotonic smoothing}, url = {http://dx.doi.org/10.1145/1242572.1242582}, year = {2007} }

TY - CONF ID - citeulike:1288655 L3 - citeulike-article-id:1288655 SN - 9781595936547 AD - New York, NY, USA EP - 70 TI - Page-level template detection via isotonic smoothing PB - ACM Press T2 - WWW '07: Proceedings of the 16th international conference on World Wide Web SP - 61 KW - classification KW - similarity KW - text N1 - Simple approach: hash fragment, check frequencies. Approach here: site-level. No training data required. The function that is learned goes from DOM nodes to {template/nontemplate}. Features: placement, bgcolor, aspect ratio, link density, avg. sentence site, etc. Smoothing: all the children of a template are templates. AU - Chakrabarti, Deepayan AU - Kumar, Ravi AU - Punera, Kunal PY - 2007/// UR - http://dx.doi.org/10.1145/1242572.1242582 ER -