Each mapper creates a support vector once the algorithm starts. As a result, the Hadoop cluster will have `N' support vectors. Each mapper parses to the next step by reading data in the form of 〈Instancei,Targeti,type〉. If the `type' field value is `train,' then the support vector...