SVM
Class BootStrapper

java.lang.Object
  SVM.BootStrapper

public class BootStrapper
extends java.lang.Object

Lets us grow the training file in a semi-supervised fashion.

Author:

davidc

Constructor Summary

BootStrapper()

Method Summary

static void	main(java.lang.String[] args)
static void	purifyResults(java.util.List<KnownDoc> testL) Supposedly this method will delete those entries that are incorrectly classified.

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

BootStrapper

public BootStrapper()

Method Detail

main

public static void main(java.lang.String[] args)
                 throws java.lang.Exception

Parameters:

args -

Throws:

java.lang.Exception

purifyResults

public static void purifyResults(java.util.List<KnownDoc> testL)

Supposedly this method will delete those entries that are incorrectly classified. This simply works by training on all known files, and then classifying these files. Since the SVM has seen all these files during training, it should be 100% correct classification (so those that aren't correctly classified now were either done incorrectly at first or are difficult/on the border)