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: 
ycenycute
‎09-29-2023
  • 66 Posts
  • 2 Likes Given
  • 0 Solutions
  • 4 Likes Received

SAM EM: confusion matrix on test/validation data?

by Obsidian | Level 7 in SAS Data Science
‎11-05-2021 11:18 PM
‎11-05-2021 11:18 PM
I have a dataset where the Y variable is a binary variable. I have partitioned the data into training and testing, and connect a logistic regression node to the partition node. I am wondering the is confusion matrix in the results only for the training data only? Cuz I see the role = TRAIN (see below). If so, is there a way to get the confusion matrix for testing data to do the evaluation?     ... View more

Re: SAS EM: minimum support percent in clustering node

by Obsidian | Level 7 in SAS Data Science
‎11-05-2021 09:53 PM
‎11-05-2021 09:53 PM
In the property panel of the association node in SAS EM, we have one parameter called support percentage, and here is the explanation of this parameter: "Specifies the minimum transaction frequency to support an association. The frequency is expressed as a frequency. The frequency is expressed as a percentage." ... View more

Re: SAS EM: cross validation

by Obsidian | Level 7 in SAS Data Science
‎11-05-2021 09:43 PM
‎11-05-2021 09:43 PM
Thanks for the links. I am aware of those posts. Wondering is there easy way to do it without writing codes? Like drag some nodes and run it? ... View more

SAS EM: cross validation

by Obsidian | Level 7 in SAS Data Science
‎11-05-2021 07:44 AM
‎11-05-2021 07:44 AM
How to perform cross validation on training data to tune hyperparameters in SAS EM? Any tutorial? Cannot find useful resources. ... View more

SAS EM: minimum support percent in clustering node

by Obsidian | Level 7 in SAS Data Science
‎11-05-2021 07:41 AM
‎11-05-2021 07:41 AM
In the property panel for the clustering node, we have support percentage, is this the minimum support percentage for selecting association rule? If yes, why in the results, we see rules below this threshold. Say if we set the support percentage as 5, in the results, we see rules below 5, some having 4.**. Do I misinterpret the meaning of support percentage?      ... View more

Re: SAS EM: results analysis of association

by Obsidian | Level 7 in SAS Data Science
‎11-02-2021 10:08 AM
‎11-02-2021 10:08 AM
Thank you for the reply. But I am still confused about statistics plot. What does it mean by 2, 3 and 4 if it is the items in the rule or if it is the chain size? Would you please provide explicit explanation? How can I know whether it is items in the rule or the chain size? ... View more

SAS EM: results analysis of association

by Obsidian | Level 7 in SAS Data Science
‎10-29-2021 09:08 AM
‎10-29-2021 09:08 AM
Got several questions about the results of the association node.   1. What do 2, 3 and 4 mean here?     2. What do blue and red represent? Or what is the color bar in the bottom?   ... View more

SAS EM: Association

by Obsidian | Level 7 in SAS Data Science
‎10-29-2021 09:06 AM
‎10-29-2021 09:06 AM
I have several questions regarding association node.    1. Does it require at least one ID variable and one Target variable? Why is that? What does ID and Target do here? 2. What does maximum items mean? I can find the descriptions in the document, but still don't understand. Can someone explain? Like what is a four-way association? Why when there is sequence variable, it is reset to 2?  Maximum Items — Use the Maximum Items property to specify the maximum size of any given item set that you want to consider in an association. For the example, the default value of 4 items indicates that up to four-way associations are performed. The minimum number of Maximum Items is 1. Permissible values are nonnegative integers. If you set Maximum Items to 1 and your data source does not contain a sequence variable, then Maximum Items is reset to 2.   ... View more

Re: SAS EM: does clustering node have elbow method to select the optim...

by Obsidian | Level 7 in SAS Data Science
‎10-25-2021 10:04 AM
‎10-25-2021 10:04 AM
Thanks for the link. This is informative. But can you explain the intuition behind all these measures, CCC, and ABC without the math? Why do we want to pick the first peak as our optimal # of clusters? I need to understand why instead of just clikcing and getting the results. ... View more

Re: SAS EM: does clustering node have elbow method to select the optim...

by Obsidian | Level 7 in SAS Data Science
‎10-24-2021 10:15 PM
‎10-24-2021 10:15 PM
Oh, thanks good to know. Can you explain more about ABC method. I got this figure. How to interpret this figure and what is the Gap on the vertical axis? ... View more

SAS EM: does clustering node have elbow method to select the optimal #...

by Obsidian | Level 7 in SAS Data Science
‎10-24-2021 11:20 AM
‎10-24-2021 11:20 AM
 In SAS EM, does clustering node have elbow method to select the optimal # of clusters, like the figure below?     I am aware of CCC plot. What does cubic clustering cutoff mean? How does it determine the optimal # of clusters based on this and why is that?  ... View more

Re: SAS EM: decision tree

by Obsidian | Level 7 in SAS Data Science
‎10-18-2021 04:02 AM
‎10-18-2021 04:02 AM
What does it mean by use in search and what does it mean by branch 2? ... View more

SAS EM: how to interpret the output results from decision tree node

by Obsidian | Level 7 in SAS Data Science
‎10-18-2021 02:20 AM
‎10-18-2021 02:20 AM
Any guidelines on how to interpret output results from decision tree node in SAS EM? What is leaf statistics, fit statistics, score ranking overlay and treemap?     ... View more

Re: SAS EM: decision tree

by Obsidian | Level 7 in SAS Data Science
‎10-18-2021 01:51 AM
‎10-18-2021 01:51 AM
What is the difference in terms of handling missing data between Decision Tree node and the interactive tree? What does it mean by "custom" tree? ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎09-29-2023 07:55 AM
Likes given to
View All
Likes from
View All
Quick links