SVM in oracle database 10g: removing the barriers to widespread adoption of support vector machines Export

@inproceedings{citeulike:5952325, abstract = {Contemporary commercial databases are placing an increased emphasis on analytic capabilities. Data mining technology has become crucial in enabling the analysis of large volumes of data. Modern data mining techniques have been shown to have high accuracy and good generalization to novel data. However, achieving results of good quality often requires high levels of user expertise. Support Vector Machines (SVM) is a powerful state-of-the-art data mining algorithm that can address problems not amenable to traditional statistical analysis. Nevertheless, its adoption remains limited due to methodological complexities, scalability challenges, and scarcity of production quality SVM implementations. This paper describes Oracle's implementation of SVM where the primary focus lies on ease of use and scalability while maintaining high performance accuracy. SVM is fully integrated within the Oracle database framework and thus can be easily leveraged in a variety of deployment scenarios.}, author = {Milenova, Boriana L. and Yarmus, Joseph S. and Campos, Marcos M.}, booktitle = {VLDB '05: Proceedings of the 31st international conference on Very large data bases}, citeulike-article-id = {5952325}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1083592.1083725}, comment = {http://www.oracle.com/technology/products/bi/odm/pdf/support\_vector\_machines\_paper\_1205.pdf}, isbn = {1-59593-154-6}, keywords = {commercial\_databases, data\_mining\_algorithms, oracle\_database, stroz-friedberg, support-vector-machines, svm, text\_mining}, location = {Trondheim, Norway}, pages = {1152--1163}, posted-at = {2009-10-16 19:23:22}, priority = {5}, publisher = {VLDB Endowment}, title = {SVM in oracle database 10g: removing the barriers to widespread adoption of support vector machines}, url = {http://portal.acm.org/citation.cfm?id=1083592.1083725}, year = {2005} }

TY - CONF ID - citeulike:5952325 L3 - citeulike-article-id:5952325 N2 - Contemporary commercial databases are placing an increased emphasis on analytic capabilities. Data mining technology has become crucial in enabling the analysis of large volumes of data. Modern data mining techniques have been shown to have high accuracy and good generalization to novel data. However, achieving results of good quality often requires high levels of user expertise. Support Vector Machines (SVM) is a powerful state-of-the-art data mining algorithm that can address problems not amenable to traditional statistical analysis. Nevertheless, its adoption remains limited due to methodological complexities, scalability challenges, and scarcity of production quality SVM implementations. This paper describes Oracle's implementation of SVM where the primary focus lies on ease of use and scalability while maintaining high performance accuracy. SVM is fully integrated within the Oracle database framework and thus can be easily leveraged in a variety of deployment scenarios. CY - Trondheim, Norway SN - 1-59593-154-6 EP - 1163 TI - SVM in oracle database 10g: removing the barriers to widespread adoption of support vector machines PB - VLDB Endowment T2 - VLDB '05: Proceedings of the 31st international conference on Very large data bases SP - 1152 KW - commercial_databases KW - data_mining_algorithms KW - oracle_database KW - stroz-friedberg KW - support-vector-machines KW - svm KW - text_mining N1 - http://www.oracle.com/technology/products/bi/odm/pdf/support_vector_machines_paper_1205.pdf AU - Milenova, Boriana AU - Yarmus, Joseph AU - Campos, Marcos PY - 2005/// UR - http://portal.acm.org/citation.cfm?id=1083592.1083725 ER -