cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Search instead for 
Did you mean: 
HJain
‎05-17-2018
  • 5 Posts
  • 1 Likes Given
  • 0 Solutions
  • 0 Likes Received

Re: How to choose the best k among many in SAS?

by Calcite | Level 5 in SAS Data Science
‎05-15-2018 06:55 PM
‎05-15-2018 06:55 PM
Hi @WendyCzika.. Thank you for your response. What I understood from the post you referred to was that I should analyse the CCC plot. I don't know if its me doing something wrong or not, but when I am  specifying the number of clusters and NOT going through Automatic way, I do not get the CCC plot.   Also, when I have calculated Intra-Cluster Similarity, how can I use to deduce the best k?   Thank you so much! ... View more

How to choose the best k among many in SAS?

by Calcite | Level 5 in SAS Data Science
‎05-15-2018 02:49 AM
‎05-15-2018 02:49 AM
Hi Guys,   I am implementing cluster modelling technique on a data set using Cluster module under Explore tab. The number of clusters are user specific in order to learn which k performs the best. I have chosen k = 5,6,7,8,9 for the same data set.   How can I actually deduce which k is the best. I have tried to calculate Intra Cluster Similarity for each cluster of each k using inter cluster distance given in the results but do not how to deduce anything fruitful.   Any help would be really appreciated.   Thank you.   ... View more

Re: What are the characteristics of a good cluster?

by Calcite | Level 5 in SAS Data Science
‎05-04-2018 11:47 PM
‎05-04-2018 11:47 PM
Thank you for the suggestion 🙂 ... View more

What are the characteristics of a good cluster?

by Calcite | Level 5 in SAS Data Science
‎05-03-2018 09:30 PM
‎05-03-2018 09:30 PM
I am analysing a data set using Clustering Algorithm in SAS Enterprise Miner. I have tried to change various settings to get the best results. But,I dont really know how to decide whether the model created is an optimal one or not. I have tried to compared it using various metrics: Number of Clusters, Frequency of Segments and CCC(Cubic Clustering Criterion). when Internal Standardization -> None Number of clusters is 43. CCC -> 20.6302 Frequency of clusters ranges from 2 to 45 when Internal Standardization -> Standardization Number of clusters is 20. CCC -> 30.55503 Frequency of clusters ranges from 3-86 when Internal Standardization -> Range standardization Number of clusters is 3. CCC -> -7.68585 Frequency of clusters ranges from 52-289 CCC Cutoff is 3. Any help would be really appreciated. Thanks. ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎05-17-2018 08:28 AM
Likes given to
View All
Quick links