Oracle Data Mining and the implementation of Support Vector Machine

Contemporary commercial databases are putting an increased accentuation on analytic abilities. The data mining technology is very important when we are referring to large volumes of data for analysis. Regarding novel data if we use modern data mining techniques we may improve their accuracy and generalization. But as we all know attaining results of good quality frequently demands high level of proficiency and user expertise. Support Vector Machines is a wonderful and potent state-of-the-art data mining algorithm and can express problems not compliant to traditional statistical analysis. Anywise, this kind of algorithm stays limited on the strength of methodological complexities, scalability challenges, and scarcity of production quality SVM implementations. The paper hereby portrays Oracle's implementation of SVM where the primary topic lies on ease of use and scalability whilst maintaining high performance accuracy. Support Vector Machines algorithm is entirely integrated into the Oracle database framework and so it can be easily leveraged in a multifariousness of deployment scenarios.

Autorentext

Victoria Cupet is a consultant, trainer and coach with an experience of more than 10 years in such fields as Business Analysis, Process Management, Project Managements and Agile.