GetAnotherLabel.DawidSkene Class Reference

Public Member Functions
	DawidSkene (List< Labeling > labels, List< Labeling > correct_labels, List< Labeling > classification_cost)
void	UpdateAnnotatorCosts ()
void	UpdateAnnotatorErrorRates ()
void	UpdateObjectClassProbabilities ()
double[]	GetObjectClassProbabilities (int oid, int lid)
void	UpdatePriors ()
double	GetAnnotatorCostNaive (int k)
	Estimates the cost for annotator k without attempting corrections of labels. More...
double	GetAnnotatorCostMinimized (int k)
	Estimates the cost for annotator k after adjusting class estimates using the error rates of the annotator. More...
double	GetAnnotatorCostAdjusted (int k)
	Estimates the cost for annotator k after adjusting class estimates using the error rates of the annotator More...
Dictionary< string, string >	GetMajorityVote ()
Dictionary< string, double[][]>	GetConfusionMatrix ()
void	Estimate (int iterations)
string	PrintAnnotatorCostsSummary ()
string	PrintAllWorkerScores ()
string	PrintWorkerScore (int workerid)
	Like printAnnotatorCosts, but doesn't print cost_naive and cost_adj More...
string	PrintObjectClassProbabilities (double entropy_threshold)
	Prints the objects that have probability distributions with entropy higher than the given threshold. More...
void	PrintObjectClassProbabilities ()
double[][]	GetObjectClassProbabilities ()
string	PrintPriors ()
string	PrintDiffVote (Dictionary< string, string > prior_voting, Dictionary< string, string > posterior_voting)
string	PrintVote ()

Static Public Member Functions
static double	Entropy (double[,] p, int i)

Public Attributes
Dictionary< string, int >	classes

Static Public Attributes
static char	splitChar = ','

Member Function Documentation

double GetAnotherLabel.DawidSkene.GetAnnotatorCostAdjusted ( int  k )

inline

Estimates the cost for annotator k after adjusting class estimates using the error rates of the annotator

Parameters

k

k

Returns

The expected cost of misclassifications of annotator k

double GetAnotherLabel.DawidSkene.GetAnnotatorCostMinimized ( int  k )

inline

Estimates the cost for annotator k after adjusting class estimates using the error rates of the annotator.

Parameters

k

k

Returns

The expected cost of misclassifications of annotator k.

double GetAnotherLabel.DawidSkene.GetAnnotatorCostNaive ( int  k )

inline

Estimates the cost for annotator k without attempting corrections of labels.

Parameters

k

k

Returns

The expected cost of misclassifications of annotator k.

string GetAnotherLabel.DawidSkene.PrintObjectClassProbabilities ( double  entropy_threshold )

inline

Prints the objects that have probability distributions with entropy higher than the given threshold.

Parameters

entropy_threshold

entropy_threshold

Returns

string GetAnotherLabel.DawidSkene.PrintWorkerScore ( int  workerid )

inline

Like printAnnotatorCosts, but doesn't print cost_naive and cost_adj

Parameters

workerid

Returns

---

The documentation for this class was generated from the following file:

• C:/Users/Matteo/Source/Repos/active-crowd/CrowdsourcingModels/DawidSkene/DawidSkene.cs