I am using SAS Text Miner 5.1 for supervised classification. I have the following questions. Can anyone please clarify??
1. It appears that the "Cluster" node can be used to extract the SVDs for training the classification model. How can I get the Term Document Matrix for further analysis???
2. Of the modeling methods available (AutoNeural, DecisionTree, Gradient Boosting, MBR, Neural Network, Rule Induction, SVM, TwoStage), which of the ones can be used for multi class classification problem (the target variable has more than 2 classes). I tried using SVM but getting an error that the target must be binary. How can I extend this to multi class classification problems???
Thank you.
Ravi
Hi Ravi,
My responses are depicted by >> to make them easier to find.
I am using SAS Text Miner 5.1 for supervised classification. I have the following questions. Can anyone please clarify??
>> Since SAS Text Miner has incorporated many changes and enhancements after version 5.1, the following response is tailored towards the newer release (version 12.1 and after.)
1. It appears that the "Cluster" node can be used to extract the SVDs for training the classification model. How can I get the Term Document Matrix for further analysis???
>> Please refer to my response to another SAS Communities post "Document Term Matrix in Text Miner" that shows you how to create a SAS data set containing the document-by-term matrix.
2. Of the modeling methods available (AutoNeural, DecisionTree, Gradient Boosting, MBR, Neural Network, Rule Induction, SVM, TwoStage), which of the ones can be used for multi class classification problem (the target variable has more than 2 classes). I tried using SVM but getting an error that the target must be binary. How can I extend this to multi class classification problems???
>> All of the modeling nodes except for the HPSVM node (which replaces the SVM node and builds an SVM model in a high performance fashion) support multi-classes classification problem. Please refer to the SAS Enterprise Miner 14.2 Reference Help for detailed information for each node.
Hope this helps!
Ann
Build your skills. Make connections. Enjoy creative freedom. Maybe change the world. Registration is now open through August 30th. Visit the SAS Hackathon homepage.
Register today!Use this tutorial as a handy guide to weigh the pros and cons of these commonly used machine learning algorithms.
Find more tutorials on the SAS Users YouTube channel.