IRPact_modellierung.perception

Class HistogramBin

java.lang.Object

IRPact_modellierung.perception.HistogramBin

public class HistogramBin
extends java.lang.Object

Class to describe the bins in a histogram as a discretization of the range / height of values in the histogram. The histogram bins associate the lambda parameter of the corresponding histogram to the strength of exponential decay (over time) of new data.

Constructor Summary

Constructors

Constructor and Description
HistogramBin(double binMinValue, double binMaxValue, ProductAttributePerceptionHistogram correspondingHistogram)

Method Summary

Modifier and Type	Method and Description
void	addValueToBin(double weight, double timestamp) adds a value to the bin based on the weight of the value, simulation time and lambda parameter of the corresponding histogram.
double	getBinHeight()
double	getBinMaxValue()
double	getBinMinValue()

Methods inherited from class java.lang.Object

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

Constructor Detail

HistogramBin

public HistogramBin(double binMinValue,
                    double binMaxValue,
                    ProductAttributePerceptionHistogram correspondingHistogram)

Parameters:

binMinValue - minimum value corresponding to the bin

binMaxValue - maximum value corresponding to the bin

correspondingHistogram - the histogram this bin is associated with

Method Detail

getBinHeight

public double getBinHeight()

getBinMinValue

public double getBinMinValue()

getBinMaxValue

public double getBinMaxValue()

addValueToBin

public void addValueToBin(double weight,
                          double timestamp)

adds a value to the bin based on the weight of the value, simulation time and lambda parameter of the corresponding histogram. New information is incorporated on an exponentially increasing / decreasing fashion (lambda > 0 / lambda <0) with the weight scaling multiplically

Parameters:

weight - The weight of the value to be added (multiplicable scaling)

timestamp - The simulation time the value is incorporated in the histogram (scales exponentially with lambda of the corresponding histogram)