Timeline for What algorithm should I use to cluster a huge binary dataset into few categories?
Current License: CC BY-SA 3.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 23, 2015 at 6:26 | comment | added | Dror Atariah | I don't have matlab either, but this is what I understood from the documentation. | |
| Apr 22, 2015 at 18:42 | comment | added | Has QUIT--Anony-Mousse | I don't know what they exactly sell by the name of "k-means". I don't have matlab. But the name implies it should be the mean and not something else. | |
| Apr 22, 2015 at 9:51 | comment | added | Dror Atariah | So, matlab uses k-means together with the hamming metric; this doesn't make much sense. | |
| Apr 21, 2015 at 14:17 | comment | added | Has QUIT--Anony-Mousse | k-means is called k-means because it uses the mean. Otherwise, it's called k-medoids, k-modes, etc. The mean is good for L2 - sum of squared deviations. | |
| Apr 21, 2015 at 10:18 | comment | added | Dror Atariah | Just to make sure I'm getting it right: matlab uses the arithmetic mean when updating the centroids when using the k-means together with the hamming metric. Is that right? What's the right way to use this metric in matlab? | |
| Apr 21, 2015 at 9:14 | comment | added | Has QUIT--Anony-Mousse | Because of the mean. Arithmetic mean is meaningless with hamming distance or binary data. Use the mode or medoid instead. | |
| Apr 20, 2015 at 12:05 | comment | added | Dror Atariah | Can you please explain why "Don't use with Hamming distance"? It might make sense, after all it is available in Matlab.I don't mind opening a new question, if it make sense. | |
| Mar 11, 2014 at 10:34 | history | answered | Has QUIT--Anony-Mousse | CC BY-SA 3.0 |