查找无向图拉普拉斯矩阵的Java程序

拉普拉斯矩阵，也称为图拉普拉斯矩阵、导纳矩阵、基尔霍夫矩阵或离散拉普拉斯矩阵，是图的矩阵表示。首先求拉普拉斯矩阵，求图的邻接矩阵和度矩阵，拉普拉斯矩阵的公式如下：

Laplacian matrix = Degree matrix – Adjacency matrix

例子

Laplacian Matrix: 
 2  -1   0   0  -1   0  
-1   3  -1   0  -1   0  
 0  -1   2  -1   0   0  
 0   0  -1   3  -1  -1  
-1  -1   0  -1   3   0  
 0   0   0  -1   0   1

基本定义：

1. 邻接矩阵：根据图顶点是否相互连接，值可以是 0 或 1。
2. 度矩阵：与顶点相邻的顶点数。

算法：

1. 添加无向图的适当边。
2. 读取图的邻接矩阵和度矩阵
3. 用公式计算拉普拉斯矩阵。

下面是上述方法的实现：

Java

// Java program to find laplacian
// matrix of an undirected graph
class Graph {
    class Edge {
        int src, dest;
    }
    int vertices, edges;
    Edge[] edge;
  
    Graph(int vertices, int edges)
    {
        this.vertices = vertices;
        this.edges = edges;
        edge = new Edge[edges];
        for (int i = 0; i < edges; i++) {
            edge[i] = new Edge();
        }
    }
    public static void main(String[] args)
    {
        int i, j;
        int numberOfVertices = 6;
        int numberOfEdges = 7;
        int[][] adjacency_matrix
            = new int[numberOfEdges][numberOfEdges];
        int[][] degree_matrix
            = new int[numberOfEdges][numberOfEdges];
        int[][] laplacian_matrix
            = new int[numberOfEdges][numberOfEdges];
        Graph g
            = new Graph(numberOfVertices, numberOfEdges);
  
        // Adding edges with source and destination
        // edge 1--2
        g.edge[0].src = 1;
        g.edge[0].dest = 2;
  
        // edge 1--5
        g.edge[1].src = 1;
        g.edge[1].dest = 5;
  
        // edge 2--3
        g.edge[2].src = 2;
        g.edge[2].dest = 3;
  
        // edge 2--5
        g.edge[3].src = 2;
        g.edge[3].dest = 5;
  
        // edge 3--4
        g.edge[4].src = 3;
        g.edge[4].dest = 4;
  
        // edge 4--6
        g.edge[5].src = 4;
        g.edge[5].dest = 6;
  
        // edge 5--4
        g.edge[6].src = 5;
        g.edge[6].dest = 4;
  
        // Adjacency Matrix
        for (i = 0; i < numberOfEdges; i++) {
            for (j = 0; j < numberOfEdges; j++) {
                adjacency_matrix[g.edge[i].src]
                                [g.edge[i].dest]
                    = 1;
                adjacency_matrix[g.edge[i].dest]
                                [g.edge[i].src]
                    = 1;
            }
        }
        System.out.println("Adjacency matrix : ");
        for (i = 1; i < adjacency_matrix.length; i++) {
            for (j = 1; j < adjacency_matrix[i].length;
                 j++) {
                System.out.print(adjacency_matrix[i][j]
                                 + " ");
            }
            System.out.println();
        }
        System.out.println("\n");
  
        // Degree Matrix
        for (i = 0; i < numberOfEdges; i++) {
            for (j = 0; j < numberOfEdges; j++) {
                degree_matrix[i][i]
                    += adjacency_matrix[i][j];
            }
        }
        System.out.println("Degree matrix : ");
        for (i = 1; i < degree_matrix.length; i++) {
            for (j = 1; j < degree_matrix[i].length; j++) {
                System.out.print(degree_matrix[i][j] + " ");
            }
            System.out.println();
        }
        System.out.println("\n");
  
        // Laplacian Matrix
        System.out.println("Laplacian matrix : ");
        for (i = 0; i < numberOfEdges; i++) {
            for (j = 0; j < numberOfEdges; j++) {
                laplacian_matrix[i][j]
                    = degree_matrix[i][j]
                      - adjacency_matrix[i][j];
            }
        }
        for (i = 1; i < laplacian_matrix.length; i++) {
            for (j = 1; j < laplacian_matrix[i].length;
                 j++) {
                System.out.printf("%2d",
                                  laplacian_matrix[i][j]);
                System.out.printf("%s", "  ");
            }
            System.out.println();
        }
    }
}

输出

Adjacency matrix : 
0 1 0 0 1 0 
1 0 1 0 1 0 
0 1 0 1 0 0 
0 0 1 0 1 1 
1 1 0 1 0 0 
0 0 0 1 0 0 


Degree matrix : 
2 0 0 0 0 0 
0 3 0 0 0 0 
0 0 2 0 0 0 
0 0 0 3 0 0 
0 0 0 0 3 0 
0 0 0 0 0 1 


Laplacian matrix : 
 2  -1   0   0  -1   0  
-1   3  -1   0  -1   0  
 0  -1   2  -1   0   0  
 0   0  -1   3  -1  -1  
-1  -1   0  -1   3   0  
 0   0   0  -1   0   1