Uses of Interface
netkit.util.Configurable

Packages that use Configurable

netkit.classifiers
netkit.classifiers.active
netkit.classifiers.nonrelational
netkit.classifiers.relational
netkit.inference

Uses of Configurable in netkit.classifiers

Subinterfaces of Configurable in netkit.classifiers

interface

Classifier $Id: Classifier.java,v 1.4 2004/12/12 17:43:40 sofmac Exp $ Part of the open-source Network Learning Toolkit User: smacskassy Date: Dec 2, 2004 Time: 9:16:44 PM

Classes in netkit.classifiers that implement Configurable

class	ClassifierImp $Id: ClassifierImp.java,v 1.6 2007/03/26 23:45:06 sofmac Exp $ Part of the open-source Network Learning Toolkit User: smacskassy Date: Dec 2, 2004 Time: 9:16:44 PM
class	NetworkLearning

Uses of Configurable in netkit.classifiers.active

Subinterfaces of Configurable in netkit.classifiers.active

interface

PickLabelStrategy

Classes in netkit.classifiers.active that implement Configurable

class	ComparatorLabeler This class does a comparison between multiple active learning strategies.
class	EmpiricalRiskMinimization
class	EmpiricalRiskMinimizationHarmonic
class	ERMHybrid This class duses multiple active learning strategies to pick the next candidate(s).
class	GraphCentralityLabeling Graph Centrality Labeling for Active Learning iteratively picks central nodes in a graph that are in clusters that have no known labels.
class	GreedyTruth Picks next labels by getting the best jump in accuracy, knowing what the truth is.
class	PickLabelStrategyImp
class	RandomLabeling
class	UncertaintyLabeling

Uses of Configurable in netkit.classifiers.nonrelational

Classes in netkit.classifiers.nonrelational that implement Configurable

class	ClassPrior This classifier is static and always returns the class marginals.
class	ExternalPrior This classifier reads in estimates from a user-specified file.
class	LocalMetaClassifier Abstract class for combining multiple classifiers.
class	LocalWeka Weka wrapper that uses a specified weka classifier to do its predictions.
class	MetaMultiplicative a classifier that multiplies the predictions of one or more classifiers and returns a normalized distribution as its own estimate.
class	NullPrior Predict nothing.
class	UniformPrior Dummy classifier that always predicts that all classes were equally likely.

Uses of Configurable in netkit.classifiers.relational

Subinterfaces of Configurable in netkit.classifiers.relational

interface

NetworkClassifier Interface for a relational classifier in addition to those of a nonrelational classifier.

Classes in netkit.classifiers.relational that implement Configurable

class	ClassDistribRelNeighbor The Class Distributional Relational Neighbor (ClassDistributRelNeighbor) classifier works by creating a 'prototypical' class vector for each class of node and then estimating a label for a new node by calculating how near that new node is to each of these 'class reference vectors'.
class	Harmonic The Harmonic Function classifier from Zhu (2003) Reference: Zhu, X., Ghahramani, Z., & Lafferty, J.
class	NetworkClassifierImp Core implementation of the NetworkClassifier (and Classifier) interface.
class	NetworkMetaClassifier Abstract class for combining multiple relational and non-relational classifiers.
class	NetworkOnlyBayes Network-only Bayes Classifier induces a naive Bayes model based on labels of neighbors of a node and uses a Markov random field formulation when one or more neighbors have estimated labels.
class	NetworkWeka Weka wrapper that uses a specified weka classifier to do its predictions.
class	ProbRelationalNeighbor This is a probablistic version of wbRN and it estimates nodes by using a Bayesian combination of the neighbors edges.
class	WeightedVoteRelationalNeighbor weighted-vote Relational Neighbor Classifier (wvRN).

Uses of Configurable in netkit.inference

Classes in netkit.inference that implement Configurable

class	GibbsSampling
class	InferenceMethod
class	IterativeClassification
class	NullInference
class	RelaxationLabeling