Data Driving Multiple Classifier System

---

Documentation for package ‘D2MCS’ version 1.0.1

• DESCRIPTION file.
• User guides, package vignettes and other documentation.
• Package NEWS.

Help Pages

Accuracy	Computes the Accuracy measure.
BinaryPlot	Plotting feature clusters following bi-class problem.
ChiSquareHeuristic	Feature-clustering based on ChiSquare method.
ClassificationOutput	D2MCS Classification Output.
ClassMajorityVoting	Implementation of Majority Voting voting.
ClassWeightedVoting	Implementation Weighted Voting scheme.
ClusterPredictions	Manages the predictions achieved on a cluster.
CombinedMetrics	Abstract class to compute the class prediction based on combination between metrics.
CombinedVoting	Implementation of Combined Voting.
ConfMatrix	Confusion matrix wrapper.
D2MCS	Data Driven Multiple Classifier System.
Dataset	Simple Dataset handler.
DatasetLoader	Dataset creation.
DefaultModelFit	Default model fitting implementation.
DependencyBasedStrategy	Clustering strategy based on dependency between features.
DependencyBasedStrategyConfiguration	Custom Strategy Configuration handler for the DependencyBasedStrategy strategy.
FisherTestHeuristic	Feature-clustering based on Fisher's Exact Test.
FN	Computes the False Negative errors.
FP	Computes the False Positive value.
GainRatioHeuristic	Feature-clustering based on GainRatio methodology.
GenericClusteringStrategy	Abstract Feature Clustering Strategy class.
GenericHeuristic	Abstract Feature Clustering heuristic object.
GenericModelFit	Abstract class for defining model fitting method.
GenericPlot	Pseudo-abstract class for creating feature clustering plots.
HDDataset	High Dimensional Dataset handler.
HDSubset	High Dimensional Subset handler.
InformationGainHeuristic	Feature-clustering based on InformationGain methodology.
Kappa	Computes the Kappa Cohen value.
KendallHeuristic	Feature-clustering based on Kendall Correlation Test.
MCC	Computes the Matthews correlation coefficient.
MCCHeuristic	Feature-clustering based on Matthews Correlation Coefficient score.
MeasureFunction	Archetype to define customized measures.
Methodology	Abstract class to compute the probability prediction based on combination between metrics.
MinimizeFN	Combined metric strategy to minimize FN errors.
MinimizeFP	Combined metric strategy to minimize FP errors.
MultinformationHeuristic	Feature-clustering based on Mutual Information Computation theory.
NoProbability	Compute performance across resamples.
NPV	Computes the Negative Predictive Value.
OddsRatioHeuristic	Feature-clustering based on Odds Ratio measure.
PearsonHeuristic	Feature-clustering based on Pearson Correlation Test.
PPV	Computes the Positive Predictive Value.
Precision	Computes the Precision Value.
PredictionOutput	Encapsulates the achieved predictions.
ProbAverageVoting	Implementation of Probabilistic Average voting.
ProbAverageWeightedVoting	Implementation of Probabilistic Average Weighted voting.
ProbBasedMethodology	Methodology to obtain the combination of the probability of different metrics.
Recall	Computes the Recall Value.
Sensitivity	Computes the Sensitivity Value.
SimpleStrategy	Simple feature clustering strategy.
SimpleVoting	Abtract class to define simple voting schemes.
SingleVoting	Manages the execution of Simple Votings.
SpearmanHeuristic	Feature-clustering based on Spearman Correlation Test.
Specificity	Computes the Specificity Value.
StrategyConfiguration	Default Strategy Configuration handler.
Subset	Classification set.
SummaryFunction	Abstract class to computing performance across resamples.
TN	Computes the True Negative value.
TP	Computes the True Positive Value.
TrainFunction	Control parameters for train stage.
TrainOutput	Stores the results achieved during training.
Trainset	Trainning set.
TwoClass	Control parameters for train stage (Bi-class problem).
TypeBasedStrategy	Feature clustering strategy.
UseProbability	Compute performance across resamples.
VotingStrategy	Voting Strategy template.