es.upm.etsisi.cf4j.data

Class RandomSplitDataSet

java.lang.Object

es.upm.etsisi.cf4j.data.RandomSplitDataSet

All Implemented Interfaces:

DataSet

public class RandomSplitDataSet
extends Object
implements DataSet

This class implements the DataSet interface by random splitting the collaborative filtering ratings allocated in a text file. Each line of the ratings file must have the following format:

<userId><separator><itemId><separator><rating>

Where <separator> is an special character that delimits ratings fields (semicolon by default).

Training and test ratings are selected randomly by the probability of an user and an item to belong to the test set.

Field Summary

Fields

Modifier and Type	Field and Description
protected static String	DEFAULT_SEPARATOR
protected ArrayList<DataSetEntry>	ratings Raw stored ratings
protected ArrayList<DataSetEntry>	testRatings Raw stored test ratings

Constructor Summary

Constructors

Constructor and Description
RandomSplitDataSet(String filename) Generates a DataSet form a text file.
RandomSplitDataSet(String filename, double testUsersPercent, double testItemsPercent) Generates a DataSet form a text file.
RandomSplitDataSet(String filename, double testUsersPercent, double testItemsPercent, long seed) Generates a DataSet form a text file.
RandomSplitDataSet(String filename, double testUsersPercent, double testItemsPercent, String separator) Generates a DataSet form a text file.
RandomSplitDataSet(String filename, double testUsersPercent, double testItemsPercent, String separator, long seed) Generates a DataSet form a text file.
RandomSplitDataSet(String filename, String separator) Generates a DataSet form a text file.

Method Summary

Modifier and Type	Method and Description
int	getNumberOfRatings() This method indicates the number of (training) ratings.
int	getNumberOfTestRatings() This method indicates the number of test ratings.
Iterator<DataSetEntry>	getRatingsIterator() This method generates an iterator to navigate through the raw ratings stored in DataSetEntries.
Iterator<DataSetEntry>	getTestRatingsIterator() This method generates an iterator to navigate through the raw test ratings stored in DataSetEntries.

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT_SEPARATOR

protected static final String DEFAULT_SEPARATOR

See Also:

Constant Field Values

ratings

protected ArrayList<DataSetEntry> ratings

Raw stored ratings

testRatings

protected ArrayList<DataSetEntry> testRatings

Raw stored test ratings

Constructor Detail

RandomSplitDataSet

public RandomSplitDataSet(String filename)
                   throws IOException

Generates a DataSet form a text file. The DataSet is loaded without test items and test users.

Parameters:

filename - File with the ratings.

Throws:

IOException - When the file is not accessible by the system with read permissions.

RandomSplitDataSet

public RandomSplitDataSet(String filename,
                          double testUsersPercent,
                          double testItemsPercent)
                   throws IOException

Generates a DataSet form a text file. The DataSet is loaded with a specific percentage of test items and test users.

Parameters:

filename - File with the ratings.

testUsersPercent - Percentage of users that will be of test.

testItemsPercent - Percentage of items that will be of test.

Throws:

IOException - When the file is not accessible by the system with read permissions.

RandomSplitDataSet

public RandomSplitDataSet(String filename,
                          double testUsersPercent,
                          double testItemsPercent,
                          long seed)
                   throws IOException

Generates a DataSet form a text file. The DataSet is loaded with a specific percentage of test items and test users. This constructor allows to define an specific random seed to ensure the reproducibility of the experiments.

Parameters:

filename - File with the ratings.

testUsersPercent - Percentage of users that will be of test.

testItemsPercent - Percentage of items that will be of test.

seed - Seed applied to the random number generator.

Throws:

IOException - When the file is not accessible by the system with read permissions.

RandomSplitDataSet

public RandomSplitDataSet(String filename,
                          double testUsersPercent,
                          double testItemsPercent,
                          String separator)
                   throws IOException

Generates a DataSet form a text file. The DataSet is loaded with a specific percentage of test items and test users.

Parameters:

filename - File with the ratings.

testUsersPercent - Percentage of users that will be of test.

testItemsPercent - Percentage of items that will be of test.

separator - Separator char between file fields.

Throws:

IOException - When the file is not accessible by the system with read permissions.

RandomSplitDataSet

public RandomSplitDataSet(String filename,
                          String separator)
                   throws IOException

Generates a DataSet form a text file. The DataSet is loaded without test items and test users.

Parameters:

filename - File with the ratings.

separator - Separator char between file fields.

Throws:

IOException - When the file is not accessible by the system with read permissions.

RandomSplitDataSet

public RandomSplitDataSet(String filename,
                          double testUsersPercent,
                          double testItemsPercent,
                          String separator,
                          long seed)
                   throws IOException

Generates a DataSet form a text file. The DataSet is loaded with a specific percentage of test items and test users. This constructor allows to define an specific random seed to ensure the reproducibility of the experiments.

Parameters:

filename - File with the ratings.

testUsersPercent - Percentage of users that will be of test.

testItemsPercent - Percentage of items that will be of test.

seed - Seed applied to the random number generator.

separator - Separator char between file fields.

Throws:

IOException - When the file is not accessible by the system with read permissions.

Method Detail

getRatingsIterator

public Iterator<DataSetEntry> getRatingsIterator()

Description copied from interface: DataSet

This method generates an iterator to navigate through the raw ratings stored in DataSetEntries.

Specified by:

getRatingsIterator in interface DataSet

Returns:

Iterator of ratings

getTestRatingsIterator

public Iterator<DataSetEntry> getTestRatingsIterator()

Description copied from interface: DataSet

This method generates an iterator to navigate through the raw test ratings stored in DataSetEntries.

Specified by:

getTestRatingsIterator in interface DataSet

Returns:

Iterator of test ratings

getNumberOfRatings

public int getNumberOfRatings()

Description copied from interface: DataSet

This method indicates the number of (training) ratings.

Specified by:

getNumberOfRatings in interface DataSet

Returns:

Number of (training) ratings

getNumberOfTestRatings

public int getNumberOfTestRatings()

Description copied from interface: DataSet

This method indicates the number of test ratings.

Specified by:

getNumberOfTestRatings in interface DataSet

Returns:

Number of test ratings