dragon.ml.seqmodel.model

Interface ModelGraph

All Known Implementing Classes:

AbstractModel, CompleteModel, GenericModel, NestedModel, NoEdgeModel

public interface ModelGraph

Interface of model graph

Copyright: Copyright (c) 2005

Company: IST, Drexel University

Version:

1.0

Author:

Davis Zhou

Method Summary

Modifier and Type	Method and Description
EdgeIterator	getEdgeIterator()
int	getEdgeNum()
int	getEndState(int index) Gets the index-th ending state.
int	getEndStateNum()
int	getLabel(int state) Gets the label of the state.
int	getLabelNum() Gets the number of unique labels.
int	getMarkovOrder() Gets the order of the markov chain.
int	getOriginalLabelNum() Gets the number of original unique labels no matter what order the markov chain is.
int	getStartState(int index) Gets the index-th starting state.
int	getStartStateNum()
int	getStateNum() Gets the number of states.
boolean	isEndState(int i) Tests if the i-th state is an ending state
boolean	isStartState(int i) Tests if the i-th state is a starting state
boolean	mapLabelToState(DataSequence seq) Maps labels to states.
boolean	mapLabelToState(DataSequence data, int len, int start) This method is used by inner model only
boolean	mapStateToLabel(DataSequence seq) Maps states to labels.

Method Detail

getStateNum

int getStateNum()

Gets the number of states. It is always equal to the number of labels except for the nested model.

Returns:

the number of states

getStartStateNum

int getStartStateNum()

Returns:

the number of states which can serve as starting states

getEndStateNum

int getEndStateNum()

Returns:

the number of states which can serve as ending states

isEndState

boolean isEndState(int i)

Tests if the i-th state is an ending state

Parameters:

i - the i-th state

Returns:

true if the i-th state is an ending state

isStartState

boolean isStartState(int i)

Tests if the i-th state is a starting state

Parameters:

i - the i-th state

Returns:

true if the i-th state is a starting state

getStartState

int getStartState(int index)

Gets the index-th starting state. The index is between zero and the number of starting states minor one

Parameters:

index - the index-th starting state

Returns:

the index-th starting state.

getEndState

int getEndState(int index)

Gets the index-th ending state. The index is between zero and the number of ending states minor one

Parameters:

index - the index-th ending state

Returns:

the index-th ending state.

getEdgeNum

int getEdgeNum()

Returns:

the number of edges

getLabel

int getLabel(int state)

Gets the label of the state. The label and the state should be the same except for nested models.

Parameters:

state - the index of the state

Returns:

the label of the state

getLabelNum

int getLabelNum()

Gets the number of unique labels. If there are 4 original labels and the markov chain is of the second order, this method returns 16.

Returns:

the number of unique labels.

getOriginalLabelNum

int getOriginalLabelNum()

Gets the number of original unique labels no matter what order the markov chain is. It is always equal to the number of labels unless markov is not the first order.

Returns:

the number of original unique labels

getMarkovOrder

int getMarkovOrder()

Gets the order of the markov chain. It is always equal to one except for the complete model.

Returns:

the order of the markov chain.

getEdgeIterator

EdgeIterator getEdgeIterator()

Returns:

the iterator for edges

mapStateToLabel

boolean mapStateToLabel(DataSequence seq)

Maps states to labels. It does nothing unless nested model

Parameters:

seq - the sequence for mapping

Returns:

true if mapping successfully

mapLabelToState

boolean mapLabelToState(DataSequence seq)

Maps labels to states. It does nothing unless nested model.

Parameters:

seq - the sequence for mapping

Returns:

true if mapping successfully

mapLabelToState

boolean mapLabelToState(DataSequence data,
                        int len,
                        int start)

This method is used by inner model only

Parameters:

data - the sequence for mapping

len - the number of tokens for mapping

start - the position of the starting token

Returns:

true if mapping successfully