Package org.decisiondeck.xmcda_oo.utils.matrix

Provides matrix interfaces and classes.

See:
Description

Interface Summary

ConstrainedMatrix.Validator	An object able to discriminate between valid and invalid values.
IAltZeroToOneMatrix	A zero-to-one matrix of alternatives.
IEvaluations	A matrix of evaluations.
IFloatMatrix<RowType,ColumnType>	A matrix which is able to store double values at a set of positions, each position being represented by a row and a column.
IRdEvaluations	A matrix of evaluations whose values can be read.
IRdFloatMatrix<RowType,ColumnType>	A matrix which is able to retrieve double values at a set of positions, each position being represented by a row and a column entries.
IRdZeroToOneMatrix<RowType,ColumnType>	A matrix which is able to retrieve values in [0, 1], i.e., doubles between 0 and 1 inclusive, at a set of positions, each position being represented by a row and a column entries.
IZeroToOneMatrix<RowType,ColumnType>	A matrix which is able to store values in [0, 1], i.e., doubles between 0 and 1 inclusive, at a set of positions, each position being represented by a row and a column.

Class Summary

AltFuzzyMatrix
ConstantCompleteMatrix<Row,Column>
ConstrainedMatrix<RowType,ColumnType>	A matrix associating double values to positions specified with a row and a column.
ConstrainedMatrix.AllValidator
ConstrainedMatrix.FuzzyValidator
EvaluationsDoubleView	An unmodifiable view of the union of two sets of evaluations.
EvaluationsView
FloatMatrix<RowType,ColumnType>
FloatMatrixRestrictedView
FuzzyMatrix<RowType,ColumnType>
OneMinusInverseMatrix<PosType>	A fuzzy matrix based on an other fuzzy matrix (called 'base' matrix).

Enum Summary

ConstrainedMatrix.ValidatorType
IAltZeroToOneMatrix.Type

Package org.decisiondeck.xmcda_oo.utils.matrix Description

Provides matrix interfaces and classes.

General comment

Please note that matrix is intended to mean that the structure may not be completely defined. There might be missing values at some position specified by (row, column) even if the given row and the given column both exist in the matrix. A row (a column) exists if and only if there is at least one value associated to a position with this row (this column) in the matrix. A relation (in the mathematical sense) is a fuzzy (or binary) matrix with the supplementary restrictions that the matrix must be square and complete. Note that being complete is different than being total as the mathematics define that word for a relation: a relation over a set A is total iff for each a, b in A, either aRb or bRA. A matrix is complete as defined in this package iff there is a value (which can be zero) for each position a, b. Thus a relation is necessary complete, meaning simply (in the case of a binary relation) that for each a, b in A, either aRb or not aRb. A matrix provides a supplementary case where aRb is null, possibly interpreted as “unknown” or “not set yet”.