Due to recent updates, all users are required to create an Altair One account to login to the RapidMiner community. Click the Register button to create your account using the same email that you have previously used to login to the RapidMiner community. This will ensure that any previously created content will be synced to your Altair One account. Once you login, you will be asked to provide a username that identifies you to other Community users. Email us at Community with questions.
Comparing kNN with SVM on Two Different Data Sets
Hello,
I want to import two data sets, specifically the iris (which is found in the samples) and the wine data set (which I can import via Read CSV) and up to here, the problem with the wine data set is that instead of reading each attribute name separetly, e.g. "Alcohol", "Ash", "Phenols" etc, it combines all attribute names together in one name. In my CSV file, they are properly separated.
Next, I want to train on both of them with the kNN and SVM models using 10-fold validation and output each models Accuracy, Precision, Recall and F-Measure. The problem I have faced for this is that I used the "Cross Validation" operator and inside its parameters I have placed the kNN model (just to test for one model) and after I use the Apply Model operator on the Performance operator, I only get "accuracy" and "kappa" in the Criterion category. What I noticed is that If I use the "Titanic Training" dataset from the samples, it correctly outputs all the performance criterion that I have mentioned above (Precision, Recall etc). So why is that? Also, could I use the Compare ROCs operator to do it instead of Cross Validation?
I am circulating between youtube videos and the tutorials of Rapidminer for the last 8 hours to no avail. All I want to do is use the wine and iris datasets to compare the kNN classifier with the SVM classifier and output Accuracy, Precision, Recall and F-Measure. I am a newbie on Rapidminer, I really want to learn how to use it properly since it seems so polished and straightforward.
I want to import two data sets, specifically the iris (which is found in the samples) and the wine data set (which I can import via Read CSV) and up to here, the problem with the wine data set is that instead of reading each attribute name separetly, e.g. "Alcohol", "Ash", "Phenols" etc, it combines all attribute names together in one name. In my CSV file, they are properly separated.
Next, I want to train on both of them with the kNN and SVM models using 10-fold validation and output each models Accuracy, Precision, Recall and F-Measure. The problem I have faced for this is that I used the "Cross Validation" operator and inside its parameters I have placed the kNN model (just to test for one model) and after I use the Apply Model operator on the Performance operator, I only get "accuracy" and "kappa" in the Criterion category. What I noticed is that If I use the "Titanic Training" dataset from the samples, it correctly outputs all the performance criterion that I have mentioned above (Precision, Recall etc). So why is that? Also, could I use the Compare ROCs operator to do it instead of Cross Validation?
I am circulating between youtube videos and the tutorials of Rapidminer for the last 8 hours to no avail. All I want to do is use the wine and iris datasets to compare the kNN classifier with the SVM classifier and output Accuracy, Precision, Recall and F-Measure. I am a newbie on Rapidminer, I really want to learn how to use it properly since it seems so polished and straightforward.
Tagged:
0
Answers