The following node is available in the Open Source KNIME predictive analytics and data mining platform version 2.7.1. Discover over 1000 other nodes, as well as enterprise functionality at http://knime.com.

NG Visual Clusterer (beta)

Allows for an interactive clustering on neighborgrams. The data structure needs to be constructed beforehand using the NG Construct node. The accompanying view has different possibilities to alter the current set of accepted clusters; it also allows for an on-the-fly modification of individual clusters.

More (general) information on the neighborgram data structure and the interactive cluster generation can be found in the description of NG Construct&View node and the respective publications.

The model output of this neighborgram is the appropriate input of the neighborgram predictor node.

Note: This node needs to be executed interactively, that means, the user must choose the set of neighborgram clusters in the view and commit his choice by clicking the "Finish" button.

Ports

Input Ports
0 Model from Neighborgram Construct node.
Output Ports
0 Model containing accepted clusters.

Views

Interactive Neighborgrams
This view displays the neighborgrams. It has various control options.
This node is contained in KNIME Neighborgrams & Parallel Universe Nodes provided by KNIME GmbH, Konstanz, Germany.