05.04.2013 Views

Self-Organizing Maps, Principal Components and Non-negative ...

Self-Organizing Maps, Principal Components and Non-negative ...

Self-Organizing Maps, Principal Components and Non-negative ...

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

<strong>Self</strong> <strong>Organizing</strong> <strong>Maps</strong><br />

<strong>Principal</strong> <strong>Components</strong>, Curves <strong>and</strong> Surfaces<br />

<strong>Non</strong>-<strong>negative</strong> Matrix Factorization<br />

mutual K-nearest neighbor graph<br />

<strong>Principal</strong> <strong>Components</strong><br />

<strong>Principal</strong> Curves<br />

Spectral Clustering<br />

NK ... a symmetric set of nearby pairs of points<br />

We connect all symmetric nearest points <strong>and</strong> give them edge<br />

weight wii ′ = sii ′ (otherwise zero)<br />

We set to zero all the pairwise similarities not in NK <strong>and</strong> draw<br />

the graph.<br />

Karoline Geissler <strong>Self</strong>-<strong>Organizing</strong> <strong>Maps</strong>, <strong>Principal</strong> <strong>Components</strong> <strong>and</strong> <strong>Non</strong>-<strong>negative</strong> M

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!