K-最近邻算法：使用JavaFX绘制决策边界

下面是一个使用JavaFX绘制K-最近邻算法决策边界的示例代码：

import javafx.application.Application;
import javafx.scene.Scene;
import javafx.scene.layout.Pane;
import javafx.scene.paint.Color;
import javafx.scene.shape.Circle;
import javafx.scene.shape.Rectangle;
import javafx.stage.Stage;

import java.util.ArrayList;
import java.util.List;
import java.util.Random;

public class KNNVisualization extends Application {

    private static final int WIDTH = 800;  // 窗口宽度
    private static final int HEIGHT = 600; // 窗口高度
    private static final int K = 3;       // K-最近邻算法的K值

    private List<Point> points;            // 数据点列表
    private List<Point> trainingSet;       // 训练集
    private List<Point> testSet;           // 测试集

    private Pane root;

    public static void main(String[] args) {
        launch(args);
    }

    @Override
    public void start(Stage primaryStage) {
        // 初始化数据
        initData();

        // 创建窗口和画布
        primaryStage.setTitle("K-最近邻算法可视化");
        root = new Pane();
        Scene scene = new Scene(root, WIDTH, HEIGHT);
        primaryStage.setScene(scene);

        // 绘制数据点和决策边界
        drawPoints();
        drawDecisionBoundary();

        primaryStage.show();
    }

    private void initData() {
        points = new ArrayList<>();
        trainingSet = new ArrayList<>();
        testSet = new ArrayList<>();

        Random random = new Random();
        // 生成100个随机数据点
        for (int i = 0; i < 100; i++) {
            double x = random.nextDouble() * WIDTH;
            double y = random.nextDouble() * HEIGHT;
            Point point = new Point(x, y);
            points.add(point);
        }

        // 将前70个点作为训练集，后30个点作为测试集
        trainingSet.addAll(points.subList(0, 70));
        testSet.addAll(points.subList(70, 100));
    }

    private void drawPoints() {
        // 绘制训练集的点
        for (Point point : trainingSet) {
            Circle circle = new Circle(point.getX(), point.getY(), 5, Color.RED);
            root.getChildren().add(circle);
        }

        // 绘制测试集的点
        for (Point point : testSet) {
            Circle circle = new Circle(point.getX(), point.getY(), 5, Color.BLUE);
            root.getChildren().add(circle);
        }
    }

    private void drawDecisionBoundary() {
        // 绘制决策边界
        for (int x = 0; x < WIDTH; x++) {
            for (int y = 0; y < HEIGHT; y++) {
                Point point = new Point(x, y);
                // 使用K-最近邻算法预测该点的类别
                String predictedClass = predictClass(point);
                Color color = predictedClass.equals("A") ? Color.GREEN : Color.YELLOW;
                Rectangle rect = new Rectangle(x, y, 1, 1);
                rect.setFill(color);
                root.getChildren().add(rect);
            }
        }
    }

    private String predictClass(Point point) {
        // 使用K-最近邻算法预测数据点的类别
        List<Point> nearestNeighbors = findNearestNeighbors(point);
        int countA = 0;
        int countB = 0;

        for (Point neighbor : nearestNeighbors) {
            if (neighbor.getType().equals("A")) {
                countA++;
            } else {
                countB++;
            }
        }

        return countA > countB ? "A" : "B";
    }

    private List<Point> findNearestNeighbors(Point point) {
        // 寻找K个最近邻数据点
        List<Point> nearestNeighbors = new ArrayList<>();
        List<Double> distances = new ArrayList<>();

        for (Point trainingPoint : trainingSet) {
            double distance = calculateDistance(point, trainingPoint);
            int index = 0;
            for (double d : distances) {
                if (distance > d) {
                    break;
                }
                index++;
            }
            distances.add(index, distance);
            nearestNeighbors.add(index, trainingPoint);
        }

        return nearestNeighbors.subList(0, K);
    }

    private double calculateDistance(Point p1, Point p2) {
        // 计算两个点之间的欧式距