dragon.ml.seqmodel.data
Class BasicDataSequence

java.lang.Object
  |
  +--dragon.ml.seqmodel.data.BasicDataSequence

All Implemented Interfaces:

DataSequence

public class BasicDataSequence

extends java.lang.Object

implements DataSequence

Basic data structure for sequence data

Copyright: Copyright (c) 2005

Company: IST, Drexel University

Version:

1.0

Author:

Davis Zhou

Constructor Summary

BasicDataSequence()

BasicDataSequence(int length)

Method Summary

void	add(BasicToken token)
DataSequence	copy() Makes a copy of the current data sequence
int	getLabel(int pos) Gets the label of the token in the given position
int	getOriginalLabel(int pos) Gets the original label of the token in the given position.
Dataset	getParent() Gets the dataset object which the current data sequence belongs to
int	getSegmentEnd(int segmentStart) Gets the ending position of the segment beginning at the given position
BasicToken	getToken(int pos) Gets the token in the given position of the sequence
int	length()
void	setLabel(int pos, int label) Sets the label to the token in the given position.
void	setParent(Dataset parent) Sets the dataset object which the current data sequence belongs to
void	setSegment(int segmentStart, int segmentEnd, int label) Marks the bondary of the segment and sets the given label to all tokens of the segment.

Methods inherited from class java.lang.Object

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

Constructor Detail

BasicDataSequence

public BasicDataSequence()

BasicDataSequence

public BasicDataSequence(int length)

Method Detail

copy

public DataSequence copy()

Description copied from interface: DataSequence

Makes a copy of the current data sequence

Specified by:

copy in interface DataSequence

Returns:

a copy of the current data sequence

getParent

public Dataset getParent()

Description copied from interface: DataSequence

Gets the dataset object which the current data sequence belongs to

Specified by:

getParent in interface DataSequence

Returns:

he dataset object which the current data sequence belongs to

setParent

public void setParent(Dataset parent)

Description copied from interface: DataSequence

Sets the dataset object which the current data sequence belongs to

Specified by:

setParent in interface DataSequence

Parameters:

parent - the parsent dataset

length

public int length()

Specified by:

length in interface DataSequence

Returns:

the length of the sequence

getLabel

public int getLabel(int pos)

Description copied from interface: DataSequence

Gets the label of the token in the given position

Specified by:

getLabel in interface DataSequence

Parameters:

pos - the position in the sequence

Returns:

the label of the token in the given position

getOriginalLabel

public int getOriginalLabel(int pos)

Description copied from interface: DataSequence

Gets the original label of the token in the given position. For high order markov models, the getLabel method returns a label which combines the current label and the lables of previous positions. However, this method still returns the original label of the given position.

Specified by:

getOriginalLabel in interface DataSequence

Parameters:

pos - the position in the sequence

Returns:

the label of the token in the given position

getToken

public BasicToken getToken(int pos)

Description copied from interface: DataSequence

Gets the token in the given position of the sequence

Specified by:

getToken in interface DataSequence

Parameters:

pos - the position of the token

Returns:

a token object in the given position

setLabel

public void setLabel(int pos,
                     int label)

Description copied from interface: DataSequence

Sets the label to the token in the given position. Same as the getLabel method, for high order markov models, the input label is actually a combined label. This method will convert the combined label to original lable (i.e. the label in the first order)

Specified by:

setLabel in interface DataSequence

Parameters:

pos - the position of the token

label - the label for the position

add

public void add(BasicToken token)

getSegmentEnd

public int getSegmentEnd(int segmentStart)

Description copied from interface: DataSequence

Gets the ending position of the segment beginning at the given position

Specified by:

getSegmentEnd in interface DataSequence

Parameters:

segmentStart - the starting position of the segment

Returns:

the ending position of the segment beginning at the given position

setSegment

public void setSegment(int segmentStart,
                       int segmentEnd,
                       int label)

Description copied from interface: DataSequence

Marks the bondary of the segment and sets the given label to all tokens of the segment.

Specified by:

setSegment in interface DataSequence

Parameters:

segmentStart - the starting position of the segment

segmentEnd - the ending position of the segment

label - the lable for the segment