🎉 🎉 RAPIDMINER 9.10 IS OUT!!! 🎉🎉

Download the latest version helping analytics teams accelerate time-to-value for streaming and IIOT use cases.

CLICK HERE TO DOWNLOAD

[SOLVED] RapidMiner Self-Organizing Map "number of dimensions" is mandatory?

ben_hben_h Member Posts: 17 Contributor II
The SOM operator provided in RapidMiner "core" seems to require me to specify the number of attributes of the resultant ExampleSet. Is this analogous to the number of clusters as in the usual terminology of SOMs? Doesn't this go against the principle of the algorithm? It's an unsupervised process, meaning it's data-driven, and arrives at a suitable number of clusters depending on the data?

Answers

  • Nils_WoehlerNils_Woehler Member Posts: 463  Maven
    Hi,

    no this is not the same as the number of clusters. The number of dimensions specifies the number of dimensions after the SOM transformation.
    E.g. if you have a data set with 5 attributes and set the number of dimensions to 2, each example is mapped to a point in the 2 dimensional space created by the SOM algorithm.

    Best,
    Nils
Sign In or Register to comment.