Is it possible to use a support vector machine in combination with agglomerative clusterer?

Question

Is it possible to use support vector machine in combination with a clustering algorithm somehow? What is a sample use-case where both of them need to communicate with each other?

Answer 1

You can always use clustering to partition your data set and learn multiple classifiers, then use ensemble methods to combine the classification results.

If a class consists of multiple clusters, this can improve accuracy to learn both sub-classes and merge them after classification.