Options

How to get a ROC curve and AUC from unsupervised algotithm?

limingliming Member Posts: 13 Contributor I
edited June 2019 in Help
My dataset and the process in the file.zip as following.
 In the anomaly detection extension,  I use k-NN Global Anomaly Score operator to get the anomaly score of every instance in the dataset. The Generate ROC operator in the extension can only get the true_positive_rate and false positive rate, however, I can't get the ROC curve and AUC. Can I use some operator to directly get the ROC curve and AUC on the basis of my present process in RapidMiner?

Answers

  • Options
    MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,517 RM Data Scientist
    Hi,
    if your data has labels you can just use the normal AUC from Performance (Binominals).
    Beset,
    Martin
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
Sign In or Register to comment.