Contextual search and name disambiguation in email using graphs Export

@inproceedings{MinkovEtAl2006:ContextSearchInEmail, abstract = {Similarity measures for text have historically been an impor- tant tool for solving information retrieval problems. In many interesting settings, however, documents are often closely connected to other documents, as well as other non-textual objects: for instance, email messages are connected to other messages via header information. In this paper we consider extended similarity metrics for documents and other objects embedded in graphs, facilitated via a lazy graph walk. We provide a detailed instantiation of this framework for email data, where content, social networks and a timeline are in- tegrated in a structural graph. The suggested framework is evaluated for two email-related problems: disambiguating names in email documents, and threading. We show that reranking schemes based on the graph-walk similarity mea- sures often outperform baseline methods, and that further improvements can be obtained by use of appropriate learn- ing methods.}, address = {New York, NY, USA}, author = {Minkov, Einat and Cohen, William W. and Ng, Andrew Y.}, booktitle = {SIGIR '06: Proceedings of the 29th annual international ACM SIGIR conference on Research and development in information retrieval}, citeulike-article-id = {1118011}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1148179}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/1148170.1148179}, doi = {10.1145/1148170.1148179}, isbn = {1595933697}, keywords = {disambiguation, email, search, sigir06}, month = {August}, pages = {27--34}, posted-at = {2007-02-23 00:56:01}, priority = {2}, publisher = {ACM Press}, title = {Contextual search and name disambiguation in email using graphs}, url = {http://dx.doi.org/10.1145/1148170.1148179}, year = {2006} }

TY - CONF ID - MinkovEtAl2006:ContextSearchInEmail L3 - citeulike-article-id:1118011 N2 - Similarity measures for text have historically been an impor- tant tool for solving information retrieval problems. In many interesting settings, however, documents are often closely connected to other documents, as well as other non-textual objects: for instance, email messages are connected to other messages via header information. In this paper we consider extended similarity metrics for documents and other objects embedded in graphs, facilitated via a lazy graph walk. We provide a detailed instantiation of this framework for email data, where content, social networks and a timeline are in- tegrated in a structural graph. The suggested framework is evaluated for two email-related problems: disambiguating names in email documents, and threading. We show that reranking schemes based on the graph-walk similarity mea- sures often outperform baseline methods, and that further improvements can be obtained by use of appropriate learn- ing methods. SN - 1595933697 AD - New York, NY, USA EP - 34 TI - Contextual search and name disambiguation in email using graphs PB - ACM Press T2 - SIGIR '06: Proceedings of the 29th annual international ACM SIGIR conference on Research and development in information retrieval SP - 27 KW - disambiguation KW - email KW - search KW - sigir06 AU - Minkov, Einat AU - Cohen, William AU - Ng, Andrew PY - 2006/08// UR - http://dx.doi.org/10.1145/1148170.1148179 ER -