Package br.unip.aps.ml

Class ClusterizacaoHotspots

java.lang.Object

br.unip.aps.ml.ClusterizacaoHotspots

public final class ClusterizacaoHotspots
extends Object

Clusterizacao geografica dos focos para identificar hotspots (aprendizado nao supervisionado).

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static final record	ClusterizacaoHotspots.Hotspot	Um hotspot (grupo de focos).
static final record	ClusterizacaoHotspots.Resultado	Resultado de uma clusterizacao.

Constructor Summary

Constructors

Constructor	Description
ClusterizacaoHotspots(long semente)

Method Summary

Modifier and Type	Method	Description
double[]	cotovelo(List<FocoIncendio> focos, int kMax)	Metodo do cotovelo: distorcao (WCSS medio por foco, km²) para k = 1..kMax.
ClusterizacaoHotspots.Resultado	dbscan(List<FocoIncendio> focos, double epsKm, int minPts)
static double	distanciaKm(double lat1, double lon1, double lat2, double lon2)	Distancia de grande circulo (formula de haversine).
ClusterizacaoHotspots.Resultado	kMeans(List<FocoIncendio> focos, int k)

Methods inherited from class java.lang.Object

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

Constructor Details

ClusterizacaoHotspots

public ClusterizacaoHotspots(long semente)

Method Details

kMeans

public ClusterizacaoHotspots.Resultado kMeans(List<FocoIncendio> focos,
 int k)

dbscan

public ClusterizacaoHotspots.Resultado dbscan(List<FocoIncendio> focos,
 double epsKm,
 int minPts)

cotovelo

public double[] cotovelo(List<FocoIncendio> focos,
 int kMax)

Metodo do cotovelo: distorcao (WCSS medio por foco, km²) para k = 1..kMax.

distanciaKm

public static double distanciaKm(double lat1,
 double lon1,
 double lat2,
 double lon2)

Distancia de grande circulo (formula de haversine).