keel.Algorithms.Rule_Learning.C45Rules

Class Algorithm

java.lang.Object

keel.Algorithms.Rule_Learning.C45Rules.Algorithm

Direct Known Subclasses:

C45

public abstract class Algorithm
extends java.lang.Object

Abstract class that contains the basic structures to implements Rule learining algorithms.

Since:

JDK1.2

Version:

1.1

Author:

Written by Antonio Alejandro Tortosa (University of Granada) 01/07/2008, Modified by Xavi Solé (La Salle, Ramón Llull University - Barcelona) 12/12/2008

Field Summary

Fields

Modifier and Type	Field and Description
protected int	correct Correctly classified itemsets.
protected static java.io.BufferedWriter	log The log file.
protected MyDataset	modelDataset The model dataset.
protected static java.lang.String	modelFileName The name of the file that contains the information to build the model.
protected static java.lang.String	resultFileName The name of the result file.
protected long	startTime The instant of starting the algorithm.
protected int	testCorrect Correctly classified in test.
protected MyDataset	testDataset The test dataset.
protected static java.lang.String	testFileName The name of the file that contains the information to make the test.
protected static java.lang.String	testOutputFileName The name of the test output file.
protected MyDataset	trainDataset The train dataset.
protected static java.lang.String	trainFileName The name of the file that contains the information to make the training.
protected static java.lang.String	trainOutputFileName The name of the train output file.

Constructor Summary

Constructors

Constructor and Description
Algorithm()

Method Summary

Modifier and Type	Method and Description
protected java.lang.String	getHeader() Function to get the name of the relation and the names, types and possible values of every attribute in a dataset.
protected boolean	getNextToken(java.io.StreamTokenizer tokenizer) Puts the tokenizer in the first token of the next line.
protected void	initTokenizer(java.io.StreamTokenizer tokenizer) Function to initialize the stream tokenizer.

Methods inherited from class java.lang.Object

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

Field Detail

modelFileName

protected static java.lang.String modelFileName

The name of the file that contains the information to build the model.

trainFileName

protected static java.lang.String trainFileName

The name of the file that contains the information to make the training.

testFileName

protected static java.lang.String testFileName

The name of the file that contains the information to make the test.

trainOutputFileName

protected static java.lang.String trainOutputFileName

The name of the train output file.

testOutputFileName

protected static java.lang.String testOutputFileName

The name of the test output file.

resultFileName

protected static java.lang.String resultFileName

The name of the result file.

correct

protected int correct

Correctly classified itemsets.

testCorrect

protected int testCorrect

Correctly classified in test.

modelDataset

protected MyDataset modelDataset

The model dataset.

trainDataset

protected MyDataset trainDataset

The train dataset.

testDataset

protected MyDataset testDataset

The test dataset.

log

protected static java.io.BufferedWriter log

The log file.

startTime

protected long startTime

The instant of starting the algorithm.

Constructor Detail

Algorithm

public Algorithm()

Method Detail

initTokenizer

protected void initTokenizer(java.io.StreamTokenizer tokenizer)

Function to initialize the stream tokenizer.

Parameters:

tokenizer - The tokenizer.

getHeader

protected java.lang.String getHeader()

Function to get the name of the relation and the names, types and possible values of every attribute in a dataset.

Returns:

The name and the attributes of the relation.

getNextToken

protected boolean getNextToken(java.io.StreamTokenizer tokenizer)
                        throws java.lang.Exception

Puts the tokenizer in the first token of the next line.

Parameters:

tokenizer - The tokenizer which reads this function.

Returns:

True if reaches the end of file. False otherwise.

Throws:

java.lang.Exception - If cannot read the tokenizer.