Feature Extraction Node

There are two ways to extract features:

A Feature Extraction node uses the Non-Negative Matrix Factorization (NMF) algorithm, to build models.

If Data Miner is connected to Oracle Database 12c, a Feature Extraction node uses the PCA and SVD algorithms, described in Singular Value Decomposition and Principal Components Analysis, to build models.


Note:

Principal Components Analysis and Singular Value Decomposition models require Oracle Database 12c.

These topics describe Feature Extraction nodes: