Recent changes to K-Meanshttp://sourceforge.net/p/python-cluster/news/2006/06/k-means/Recent changes to K-MeansenTue, 27 Jun 2006 09:54:48 -0000K-Meanshttp://sourceforge.net/p/python-cluster/news/2006/06/k-means/<div class="markdown_content"><p>Late last night I started having a look into the K-Means algorithm. Seems easy as such. But.... as I want to keep a general approach so one could cluster any object it becomes a bit less a trivial task.</p>
<p>The problem/assumption of the K-Means algorithm is that the data-elements need to be representable in vector-space. This is something I cannot get around.</p>
<p>It does not look too complicated. I just have to find a general approach how to calculate the centroid of a set of objects. Maybe this requires that (similar to the distance function with the hierarchical cluster) the user needs to supply a utility function. But I'll try to avoid that.</p>
<p>Nevertheless, for objects to be clustered using the K-Means algorithm, they need to fulfill certain prerequisites. For one they need to be representable in vector-space.</p>
<p>I still need to investigate a bit on this.</p></div>Michel AlbertTue, 27 Jun 2006 09:54:48 -0000http://sourceforge.net2aa6b512a6776b0f5df4b261a80e1e275095ee5d