BookmarkSubscribeRSS Feed
ifb10
Obsidian | Level 7

Hello,

 

I have tried to implement DBSCAN and KNN clustering methods in SAS 9.4. After checking online resources, I found multiple articles on these specific clustering algorithms for SAS Viya but not much for SAS 9.4. Is there anything you could suggest? Currently, I am thinking of using PROC IML to write the algorithm myself. If there is anything ready to use, I would like to try that first.

 

Thank you!

5 REPLIES 5
PaigeMiller
Diamond | Level 26
sbxkoenk
SAS Super FREQ

DBSCAN is available in SAS 9.4 in the IMSTAT Procedure (Analytics) and uses the SAS LASR Analytic Server.

IMSTAT = In-Memory Statistics.

 

SAS LASR Analytic Server
IMSTAT Procedure (Analytics)
CLUSTER Statement
https://go.documentation.sas.com/doc/en/bicdc/9.4/inmsref/p08w4hxeqsgklrn14aqs76gs01te.htm

 

BR, Koen

ifb10
Obsidian | Level 7

Thank you Koen,

 

I have initially also tried to use IMSTAT, however I always get following error:  ERROR: Procedure IMSTAT not found.

This made me assume, that IMSTAT is not compatible with SAS 9.4, it might be an OnDemand issue however.

 

Best regards,

Tamás

Ksharp
Super User

PROC MODECLUS is for KNN.

 

Ksharp_0-1731553082450.png

 

sbxkoenk
SAS Super FREQ

MODECLUS indeed. 
Thanks for bringing this procedure from the back of my mind ... back to the front.

 

BR, Koen

sas-innovate-white.png

Our biggest data and AI event of the year.

Don’t miss the livestream kicking off May 7. It’s free. It’s easy. And it’s the best seat in the house.

Join us virtually with our complimentary SAS Innovate Digital Pass. Watch live or on-demand in multiple languages, with translations available to help you get the most out of every session.

 

Register now!

What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 5 replies
  • 1066 views
  • 4 likes
  • 4 in conversation