Regression models are predictive models. The difference between regression and classification is that regression has continuous (floating-point) target attributes, whereas classification has with discrete or categorical target attributes. In other words, if the target attribute contains continuous values, a regression technique is required.
Oracle Data Miner includes wizards that build, test, and apply regression models.
The Support Vector Machine (SVM) algorithm can be used to build regression models (as well as classification models); Oracle Data Mining uses SVM for regression models. For a breif description of the SVM algorithm, see Support Vector Machine Algorithm.
Copyright © 2006, 2008, Oracle. All rights reserved.