NearestNeighbour Class Reference

Klasse zur Erkennung des Figurtyps mithilfe des k-NearestNeighbour Verfahrens. More...

Inheritance diagram for NearestNeighbour:

Collaboration diagram for NearestNeighbour:

Public Member Functions
	NearestNeighbour (Dataset dataset)
	Constructor. More...
	NearestNeighbour (Dataset dataset, int kNeighbours)
	Constructor. More...
void	SetKNeighbours (int kNeighbour)
	Set amount of selected nearest neighbour. More...
Figure	classify (IDataPoint dataPoint)
	Classify new Figure with k- Nearest Neighbour Algorithmn. More...

Private Member Functions
double	calcEuklidDistance (IDataPoint actDataPoint, IDataPoint actPointToCompare)
	Calc euclidean distance between two Figures. More...
boolean	validateAmountOfTrainingData ()
	Check amount of Trainingsdata to config k- Parameter. More...

Private Attributes
Dataset	dataSet
int	kNeighbours = 0

Detailed Description

Klasse zur Erkennung des Figurtyps mithilfe des k-NearestNeighbour Verfahrens.

Constructor & Destructor Documentation

NearestNeighbour

(

Dataset

dataset

)

Constructor.

Parameters

dataset

Set dataset of classified Figures

NearestNeighbour	(	Dataset	dataset,
		int	kNeighbours
	)

Constructor.

Parameters

dataset	Set dataset of classified Figures
kNeighbours	Amount of selected nearest neighbours

References NearestNeighbour.kNeighbours.

Member Function Documentation

double calcEuklidDistance ( IDataPoint  actDataPoint, IDataPoint  actPointToCompare  )

private

Calc euclidean distance between two Figures.

Parameters

actDataPoint
actPointToCompare

Returns

euclidean Distance between this Figures

References IDataPoint.getArea(), and IDataPoint.getPerimeter().

Referenced by NearestNeighbour.classify().

Here is the call graph for this function:

Here is the caller graph for this function:

Figure classify

(

IDataPoint

dataPoint

)

Classify new Figure with k- Nearest Neighbour Algorithmn.

Parameters

dataPoint

Figure to classify

Implements INearestNeighbour.

References NearestNeighbour.calcEuklidDistance(), Figure.circle, Dataset.getAllData(), NearestNeighbour.kNeighbours, Logger.log(), Figure.rectangle, IDataPoint.setFigure(), Figure.UNKNOWN, and NearestNeighbour.validateAmountOfTrainingData().

Referenced by Recognition.recognizeKNN().

Here is the call graph for this function:

Here is the caller graph for this function:

void SetKNeighbours

(

int

kNeighbour

)

Set amount of selected nearest neighbour.

Parameters

Implements INearestNeighbour.

boolean validateAmountOfTrainingData ( )

private

Check amount of Trainingsdata to config k- Parameter.

If no k available, the algorithmn use 50% of trainingsdata (odd figure)

Returns

References Dataset.getAllData(), NearestNeighbour.kNeighbours, and Logger.log().

Referenced by NearestNeighbour.classify().

Here is the call graph for this function:

Here is the caller graph for this function:

Member Data Documentation

Dataset dataSet

private

int kNeighbours = 0

private

Referenced by NearestNeighbour.classify(), NearestNeighbour.NearestNeighbour(), and NearestNeighbour.validateAmountOfTrainingData().

---

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

• leImRo/src/leImRo/NearestNeighbour.java