Workflow Entry: X-Validation with One-Class SVM

Created at: 20/10/10 @ 08:37:14      Last updated: 20/10/10 @ 08:37:15
Information Version 1 (of 1)

Version created on: 20/10/10 @ 08:37:14 by: Mstolpe

Title: X-Validation with One-Class SVM

Type: RapidMiner


Information Preview

(Click on the image to get the full size)

Medium


Information Description

With this example process, I intend to show how one might validate predictions done with a one class classifier. RapidMiner includes the One-Class SVM (part of libsvm) as introduced by Schoelkopf. I'm very interested in feedback concerning problems with or errors in this experiment.

Note that the data set (Sonar) is just a toy data set chosen for demonstration - learning a one class classifier on it won't give you any good results!

DESCRIPTION The basic idea is to partition the data set as usual and to train the one class classifier only on one class inside the cross validation, but to test it on both classes for the part that was left out for testing. So for training, we have to remove all examples with the label we don't want to train the classifier on (Filter). As the SVM operator expects the nominal label values to consist of only one value, we need to map the nominal value "Mine" to "Rock". When we apply the one class model to our test data, we get "inside/outside" as a prediction. These values have to be mapped back to the original corresponding nominal values "Rock" and "Mine". Afterwards, we can use the standard performance operator.

WARNING This process will only run with a newer version of RapidMiner, because the output of the One-Class SVM - inside/out for prediction(class) - recently has been changed. In earlier versions, one has to translate the resulting confidence values for the single prediction class back to two classes.


Information Download



Information Run

Not available


Information Workflow Components

Inputs (0)
Operators (9)
Outputs (0)

Information Workflow Type

RapidMiner

Information Original Uploader

Information License

All versions of this Workflow are licensed under:

CC0 1.0 Universal (Public Domain License)

Information Credits (0)

(People/Groups)

None

Information Attributions (0)

(Workflows/Files)

None

Information Tags (0)

None

Log in to add Tags

Information Shared with Groups (0)

None

Information Featured In Packs (0)

None

Log in to add to one of your Packs

Information Ratings (0)

Current:

0.0 / 5

(0 ratings)

Log in to rate and see breakdown of ratings

Information Attributed By (0)

(Workflows/Files)

None

Information Favourited By (0)

No one

 

Citations (0)

None


Version History

In chronological order:



Reviews Reviews (0)

No reviews yet

Be the first to review!



Comments Comments (0)

No comments yet

Log in to make a comment




Workflow Other workflows that use similar services (0)

There are no workflows in myExperiment that use similar services to this Workflow.

What is this?

Linked Data

Non-Information Resource URI: http://www.myexperiment.org/workflows/1562


Alternative Formats

HTML
RDF
XML

Front Page

Home

Invite people to myExperiment

Help pages

About Us

News and Events

Mailing List

Contact Us

Developers

Publications


Taverna Workflow Workbench

myGrid

BioCatalogue

Trident

Google Coop Search

EPSRC

JISC

Microsoft

Powered by:

Rails

Icons:
Silk icon set 1.3