Feature extraction creates a set of features based on the original data. A feature is a combination of attributes in the data that is of special interest and captures important characteristics of the data. It becomes a new attribute. Typically, there are far fewer features than there are original attributes.
Applications of feature extraction include latent semantic analysis, data compression, data decomposition and projection, and pattern recognition. Feature extraction can also be used to enhance the speed and effectiveness of supervised learning by building a model using a smaller number of attributes.
Oracle Data Mining feature extraction models use the Non-Negative Matrix Factorization (NMF) algorithm. For a brief description of NMF, see Non-Negative Matrix Factorization Algorithm. NMF supports data with mining type text
.
Copyright © 2006, 2008, Oracle. All rights reserved.