Clustering with outliers (DBSCAN)

Moderator
943 Views
Highlighted
Moderator
Posts: 52
Registered: ‎08-10-2016

Clustering with outliers (DBSCAN)

[ Edited ]

dbscan clusteringdbscan clustering

Clustering enables you to find similarity groups in your data, using the well-known density-based spatial clustering of applications with noise (DBSCAN). Unlike many other clustering algorithms, DBSCAN also finds outliers. Settings for the visual let you control and refine algorithm parameters to meet your needs.

 

 


Service prerequisites: This R-powered custom visual is used without any modifications in the Power BI service

Desktop prerequisites: To run R scripts in Power BI Desktop, you must separately install R on your local computer.
You can download and install R for free from the Revolution Open download page or the CRAN Repository. For more information, see this article.

R package dependencies (see attached script for int): scales, fpc, car, dbscan

Supports R versions: R 3.3.2, R 3.3.1, R 3.3.0, MRO 3.3.2, MRO 3.3.1, MRO 3.3.0

 

 

 

 

 

 

Attachment
Attachment
Attachment
Frequent Visitor
Posts: 5
Registered: ‎03-14-2017

Re: Clustering with outliers (DBSCAN)

Thank for your post.

 

Do you know by any chance how you can identify the outliers?

 

Meaning how do I find them in my data?

 

Thank you

Moderator
Posts: 52
Registered: ‎08-10-2016

Re: Clustering with outliers (DBSCAN)

Hi, 

Unfortunatly currently we do not export data from visual.  You may need to run R-in-Power Query for that.