prea.data.preparation
Class PredefinedSplit

java.lang.Object
  prea.data.preparation.DatasetManager
      prea.data.preparation.PredefinedSplit

public class PredefinedSplit
extends DatasetManager

When a predefined split file is available, this class helps to split train/test set as defined in the file. This can be used for verifying implementation of new CF algorithm.

Since:

2012. 3. 26

Version:

1.1

Author:

Joonseok Lee

Field Summary

Fields inherited from class prea.data.preparation.DatasetManager

itemCount, itemRateAverage, K_FOLD_CROSS_VALIDATION, maxValue, minValue, PREDEFINED_SPLIT, rateMatrix, SIMPLE_SPLIT, testMatrix, userCount, userRateAverage

Constructor Summary

PredefinedSplit(SparseMatrix originalMatrix, java.lang.String splitFileName, int max, int min)
Construct an instance for splitter with predefined split file.

Method Summary

private void

readSplitData(java.lang.String fileName) Split the rating matrix into train and test set, by given split data file.

Methods inherited from class prea.data.preparation.DatasetManager

calculateAverage, getItemRateAverage, getTestMatrix, getUserRateAverage, recoverTestItems

Methods inherited from class java.lang.Object

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

Constructor Detail

PredefinedSplit

public PredefinedSplit(SparseMatrix originalMatrix,
                       java.lang.String splitFileName,
                       int max,
                       int min)

Construct an instance for splitter with predefined split file.

Method Detail

readSplitData

private void readSplitData(java.lang.String fileName)

Split the rating matrix into train and test set, by given split data file.

Parameters:

fileName - the name of split data file.