Timeline for What is an acceptable value of the Calinski & Harabasz (CH) criterion?
Current License: CC BY-SA 4.0
20 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 24, 2023 at 21:27 | comment | added | maciejwww |
Okay. So, CH is not for clustering-clustering comparison in terms of difference between them, but rather to compare how well the clusterings were created, am I right? What does it come from? What characteristics of CH provide it? And also - why can't clusters elements be strings for example (I was using sklearn implementation).
|
|
| Jun 19, 2023 at 6:01 | comment | added | ttnphns | @maciejwww, CH is an internal, not external clustering criterion. It does not compare clusters, it helps select a "better" cluster solution from several ones. | |
| Jun 19, 2023 at 5:53 | comment | added | ttnphns | @maciejwww, you may post a question, and give a data example (together with your code and calculations, if needed). | |
| Jun 18, 2023 at 20:08 | comment | added | maciejwww | @ttnphns additional questio: can u compare clusters of string with this method? | |
| Jun 18, 2023 at 20:07 | comment | added | maciejwww |
@ttnphns I'm trying to understand the method and I get the concept "the higher the better" but when I have perfect matching clusters, I get 1.0 as a reseult (with few changes it can be even 2400. How is that? Here's a link with my code and results: link.
|
|
| Dec 23, 2019 at 18:19 | comment | added | ttnphns | @MarineGalantin, (1) There is no a conceptual or other link between C-index and Silhouete index or Calinski-Harabasz, they are quite different, alternative clustering criteria to use. (2) My document does not explain how to code, rather, the .sps file attached contains the code itself, and that is SPSS syntax. If youare not SPSS user but a Python user, search for Python code elsewhere, you probably will find. However, since my doc. shows and explains the formulas, one may be able to code themselves in any language they prefer. | |
| Dec 23, 2019 at 15:50 | comment | added | Marine Galantin | and sorry to ask that additionaly, the zip is really cool, with a lot of knowledge, but it doesn't tell how to code it ? there isn't any sklearn module for that right? | |
| Dec 23, 2019 at 15:46 | comment | added | Marine Galantin | Thank you for the answer, small precision please : Is it useful ? Do you explain in that paper the link between C index and Silhouette ? | |
| Dec 23, 2019 at 15:31 | comment | added | ttnphns | @MarineGalantin, please go to my web-page and download to read description doc in collection "Internal clustering criteria". You'll find there info on C-index criterion. (This C-index should not be confused with "C-index" often used with logistic regression and ROC curves.) | |
| Dec 23, 2019 at 12:28 | comment | added | Marine Galantin | hello, thanks for the answer. What is C-index ? I can't find much about it in python. | |
| Jul 15, 2019 at 23:01 | history | edited | Nick Cox | CC BY-SA 4.0 |
added 19 characters in body
|
| Sep 21, 2018 at 21:09 | history | edited | ttnphns | CC BY-SA 4.0 |
link added
|
| Jun 27, 2018 at 14:34 | comment | added | ttnphns | I have implemented for SPSS a number of most popular clustering validation criterions, - please visit my web-page, collection "Clustering criterions". | |
| Jun 27, 2018 at 14:28 | history | edited | ttnphns | CC BY-SA 4.0 |
added 697 characters in body
|
| Apr 13, 2017 at 12:44 | history | edited | CommunityBot |
replaced http://stats.stackexchange.com/ with https://stats.stackexchange.com/
|
|
| Oct 25, 2016 at 19:42 | comment | added | ttnphns | A reader might want to see question stats.stackexchange.com/q/242360/3277 too. | |
| Apr 14, 2013 at 7:59 | history | edited | ttnphns | CC BY-SA 3.0 |
added 247 characters in body
|
| Mar 21, 2013 at 19:41 | vote | accept | greg121 | ||
| Mar 20, 2013 at 23:48 | history | edited | ttnphns | CC BY-SA 3.0 |
edited body
|
| Mar 20, 2013 at 21:05 | history | answered | ttnphns | CC BY-SA 3.0 |