SVM Class Prediction

The SVM module builds a classifier using the Support Vector Machines (SVM) class prediction method, tests a previously generated SVM classifier, and/or classifies unknown samples using a previously generated SVM classifier.

Before you begin

Generally, you use one data set to train the classifier and the other to test it. Each gene expression data set consists of two files:

GCT or RES file that contains gene expression data.
Example file (training data): all_aml_train.gct.
Example file (test data): all_aml_test.gct.
CLS file that identifies the class of each sample in the gene expression data.
Example file (training data): all_aml_train.cls.
Example file (test data): all_aml_test.cls.

The data sets must contain the same genes or SVM displays an error message.

learn more:
file formats

Step 1: SVM

The SVM module builds and/or tests a classifer by running the SVM class prediction method:

To build a classifier, specify the training data set. The module creates a classifier (*.model).
To test a previously built classifier, specify the classifier (*.model) and the test data set. The module creates a prediction results file (*.pred.odf) that assesses the accuracy of the predictor.
To build and test a classifier, specify both the training and test data sets. The module creates a classifier and a prediction results file.

learn more:
SVM

Step 2: View results

To view the prediction results file (*.pred.odf), use the PredictionResultsViewer module. For each sample, the viewer lists the actual class, predicted class, and prediction error rates.

The classifier (*.model) is a binary (machine-readable) file. It cannot be viewed, but can be used as input to the SVM module.

Considerations

The PredictionResultsViewer provides an absolute error rate (incorrect cases/total cases) and an ROC error rate (fraction of true positives versus the fraction of false positives). Use the ROC error rate for comparing results across data sets.

learn more:
PredictionResultsViewer

Step 3: Determine the class of an unknown sample

To classify unknown samples using the SVM module:

Use the saved model filename parameter to specify a previously generated classifier (*.model file).
Use the test filename parameter to specify an expression data set that contains the unknown samples.
The test class filename is a required parameter that specifies the class of each sample in the expression data set. For the unknown samples, create a class file that assigns some class (for example, "unknown") to each sample.

The module uses the classifier to predict the class of each unknown sample and creates a prediction results file. Use the PredictionResultsViewer module to view the prediction results (*.pred.odf) file:

The viewer lists each sample with its actual and predicted class.
Ignore the actual class, which was unknown.
Ignore the error rates, which are evaluating the class predictor against "known" data.