edu.umass.cs.surveyman.analyses

Class Simulation

java.lang.Object

edu.umass.cs.surveyman.analyses.Simulation

public class Simulation
extends Object

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	Simulation.ROC

Field Summary

Fields

Modifier and Type	Field and Description
static boolean	smoothing

Constructor Summary

Constructors

Constructor and Description
Simulation()

Method Summary

Methods

Modifier and Type	Method and Description
static Simulation.ROC	analyze(Survey survey, List<? extends SurveyResponse> surveyResponses, Classifier classifier, double alpha) Classifies bad actors and returns classification results for the mix of respondents provided.
static List<SurveyResponse>	simulate(Survey survey, int totalResponses, double percentAdversaries, RandomRespondent.AdversaryType adversaryType, AbstractRespondent profile) Simulates responses to a survey.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

smoothing

public static boolean smoothing

Constructor Detail

Simulation

public Simulation()

Method Detail

simulate

public static List<SurveyResponse> simulate(Survey survey,
                            int totalResponses,
                            double percentAdversaries,
                            RandomRespondent.AdversaryType adversaryType,
                            AbstractRespondent profile)
                                     throws SurveyException

Simulates responses to a survey.

Parameters:

survey - The survey to simulate.

totalResponses - The total number of responses we want to simulate.

percentAdversaries - The percentage of random actors we want to simulate.

adversaryType - The type of adversary we want to test against.

profile - An instance of the type of honest respondent we want to simulate.

Returns:

List of simulated survey responses.

Throws:

SurveyException

analyze

public static Simulation.ROC analyze(Survey survey,
                     List<? extends SurveyResponse> surveyResponses,
                     Classifier classifier,
                     double alpha)
                              throws SurveyException

Classifies bad actors and returns classification results for the mix of respondents provided.

Parameters:

survey - The survey we wish to simulate.

surveyResponses - The set of responses we wish to analyze. This will typically be a mix of some simulated responses.

classifier - The classification method to use.

alpha - A double between 0 and 1 used for various purposes by the classifiers. Sometimes alpha is used to define a confidence region. Sometimes it is used to define some small difference between real numers. Its exact meaning is determined by the classification method indicated by the classifier argument.

Returns:

A struct containing the classification results for this mix of bad actors and honest respondents.

Throws:

SurveyException