zoukankan      html  css  js  c++  java
  • Topological Sorting

    Topological sorting/ ordering is a linear ordering of its vertices such that for every directed edge uv from vertex u to vertex vu comes before v in the ordering. We can conduct topological sorting only on DAG (directed acyclic graph). Therefore, topological sorting is often used in scheduling a sequence of jobs or tasks based on their dependencies. Another typical usage is to check whether a graph has circle.

    Here is a more detailed description -> Wiki Topological sorting

    Below are several algorithms to implement topological sorting. Mostly, the time complexity is O(|V| + |E|).

    Algorithm 1 -- Ordinary Way

    Time complexity O(|V|^2 + |E|) In Step 1, time complexity O(|V|) for each vertex.

    1. Identify vertices that have no incoming edge (If no such edges, graph has cycles (cyclic graph))
    2. Select one such vertex
    3. Delete this vertex of in-degree 0 and all its outgoing edges from the graph. Place it in the output.
    4. Repeat Steps 1 to Step 3 until graph is empty

    (referrence: cs.washington Topological Sorting)

    Algorithm 2 -- Queue

    Improve algorithm 1 by use queue to store vertex whose in-degree is 0. Time complex O(|V| + |E|).

    Algorithm 3 -- DFS

    When we conduct DFS on graph, we find that if we order vertices according to its complete time, the result if one topological sorting solution.

    Time complexity O(|V| + |E|)

     1 import java.util.*;
     2 
     3 public class TopologicalSort {
     4 
     5   // When we use iteration to implement DFS, we only need 2 status for each vertex
     6   static void dfs(List<Integer>[] graph, boolean[] used, List<Integer> res, int u) {
     7     used[u] = true;
     8     for (int v : graph[u])
     9       if (!used[v])
    10         dfs(graph, used, res, v);
    11     res.add(u);
    12   }
    13
    14   public static List<Integer> topologicalSort(List<Integer>[] graph) {
    15     int n = graph.length;
    16     boolean[] used = new boolean[n];
    17     List<Integer> res = new ArrayList<>();
    18     for (int i = 0; i < n; i++)
    19       if (!used[i])
    20         dfs(graph, used, res, i);
    21     Collections.reverse(res);
    22     return res;
    23   }
    24 }

    (referrence: DFS: Topological sorting)

  • 相关阅读:
    LRU算法实现 最近最久未使用
    快速排序
    redis 常用命令
    实例:接口并发限流RateLimiter
    Hadoop伪分布式安装
    ssl免密登录(centos6)
    Linux下安装jdk-7u67-linux-x64.rpm
    Mosquitto安装和使用
    MQTT linux centOS7 部署
    CentOS更换yum源
  • 原文地址:https://www.cnblogs.com/ireneyanglan/p/4837107.html
Copyright © 2011-2022 走看看