Timeline for Unsupervised Anomaly Detection with groups
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 10, 2020 at 13:47 | comment | added | Single Malt | Yes. Each observation can have an anomaly score calculated, with values close to one more likely to be anomalies. By selecting these potential anomalies they can then be run through the trees and see how they are so easily (relatively) separated. | |
| Sep 10, 2020 at 13:45 | history | edited | Single Malt | CC BY-SA 4.0 |
added 238 characters in body
|
| Sep 10, 2020 at 12:53 | comment | added | Rogier | Hi, thanks for pointing me in the direction of one of the techniques. Although i think it could work to find outliers, I'm not quite sure how it could be used to give insights in the behaviors. Do you have any thoughts on that? | |
| Sep 9, 2020 at 14:37 | history | answered | Single Malt | CC BY-SA 4.0 |