Timeline for Can k-means be used for non normally distributed data?
Current License: CC BY-SA 4.0
19 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 8, 2019 at 13:32 | review | Suggested edits | |||
| Sep 8, 2019 at 17:23 | |||||
| Sep 3, 2019 at 16:10 | comment | added | user | "To construct the contaminated sets, the magnitudes of errors added are informed by the range of values in each dataset. For Iris, the errors added to each feature are integer values sampled from the set {𝑟: 3 ≤ 𝑟 ≤ 5}" page 5 in the paper. | |
| S Sep 3, 2019 at 11:12 | history | suggested | smci | CC BY-SA 4.0 |
more accurate version of the OP
|
| Sep 2, 2019 at 22:10 | review | Suggested edits | |||
| S Sep 3, 2019 at 11:12 | |||||
| Sep 2, 2019 at 21:10 | comment | added | smci |
Well, did you check what % of inliers in iris dataset actually lie within 3 s.d. of the centroids? It likely still happens to be true, it just doesn't automatically follow if the distribution isn't normal. Presumably the authors just need to add a one-liner clarifying that.
|
|
| Sep 2, 2019 at 14:23 | answer | added | Stephan Kolassa | timeline score: 13 | |
| Sep 2, 2019 at 14:14 | comment | added | user | My question related to the experimental of iris data set in the same paper, As I notice that iris data set not normally distributed.kaggle.com/saurabh00007/iriscsv | |
| Sep 2, 2019 at 14:11 | comment | added | user | the paper is published in IEEE | |
| S Sep 2, 2019 at 14:09 | history | edited | gung - Reinstate Monica | CC BY-SA 4.0 |
adjusted link labels
|
| S Sep 2, 2019 at 14:09 | history | suggested | Oleg Lobachev | CC BY-SA 4.0 |
adjusted link labels
|
| Sep 2, 2019 at 8:31 | review | Suggested edits | |||
| S Sep 2, 2019 at 14:09 | |||||
| Sep 2, 2019 at 7:58 | comment | added | CloseToC | The claim you quote from the paper is preceded by the assumption that the data is normal. What's needlessly restrictive in that paper is the claim that k-means assumes normality, suggesting that it couldn't be a satisfactory clustering procedure if the data isn't jointly normal. | |
| Sep 2, 2019 at 7:04 | history | became hot network question | |||
| Sep 2, 2019 at 5:51 | comment | added | Has QUIT--Anony-Mousse | What if this simply is a quite bad paper? Does not sound like a high class venue to me. | |
| Sep 2, 2019 at 0:00 | history | tweeted | twitter.com/StackStats/status/1168312739665784833 | ||
| Sep 1, 2019 at 23:57 | history | edited | user | CC BY-SA 4.0 |
added 70 characters in body
|
| Sep 1, 2019 at 23:48 | answer | added | CaucM | timeline score: 8 | |
| Sep 1, 2019 at 23:10 | history | edited | user | CC BY-SA 4.0 |
edited title
|
| Sep 1, 2019 at 22:56 | history | asked | user | CC BY-SA 4.0 |