Timeline for Can I boxcox transform variables before k-means clustering?
Current License: CC BY-SA 3.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 17, 2016 at 12:12 | comment | added | Has QUIT--Anony-Mousse | Yes, but you should be aware of the effects of PCA - the change to the result may be desired, or not (very often: not desirable). The input data should be continuous - the results of doing PCA on binary variables and then running k-means is questionable, even if it should work. | |
| Aug 17, 2016 at 11:09 | vote | accept | Ashok K Harnal | ||
| Aug 17, 2016 at 11:09 | comment | added | Ashok K Harnal | May I ask a related question or shall I put it as a new question: PCA is known to carry out ellipsoidal transformation before rendering principal components. Can I PCA before kmeans? | |
| Aug 14, 2016 at 7:37 | history | answered | Has QUIT--Anony-Mousse | CC BY-SA 3.0 |