For a undirected graph with tree characteristics, we can choose any node as the root. The result graph is then a rooted tree. Among all possible rooted trees, those with minimum height are called minimum height trees (MHTs). Given such a graph, write a function to find all the MHTs and return a list of their root labels.
Format
The graph contains n nodes which are labeled from 0 to n - 1. You will be given the number n and a list of undirected edges (each edge is a pair of labels).
You can assume that no duplicate edges will appear in edges. Since all edges are undirected, [0, 1] is the same as [1, 0] and thus will not appear together in edges.
Example 1:
Given n = 4, edges = [[1, 0], [1, 2], [1, 3]]
0
|
1
/ \
2 3
return [1]
Example 2:
Given n = 6, edges = [[0, 3], [1, 3], [2, 3], [4, 3], [5, 4]]
0 1 2
\ | /
3
|
4
|
5
return [3, 4]
Show Hint
How many MHTs can a graph have at most?
Note:
(1) According to the definition of tree on Wikipedia: “a tree is an undirected graph in which any two vertices are connected by exactly one path. In other words, any connected graph without simple cycles is a tree.”
(2) The height of a rooted tree is the number of edges on the longest downward path between the root and a leaf.
s思路:
1. 提示很重要。不看的时候,只想到了简单粗暴的方法。能不能先遍历,找到最长的path的长度,如果长度为奇数,则只有一个root,即:path的中点是minimum height tree的root;如果长度是偶数,则有两个,在path中间。
2. 如何快速找到最长的path呢?需要每个位置为root分别遍历吗?如果这样的话,复杂度就是o(V(V+E))。但,潜意识感觉这样有很多重复运算。
3. 刚想了一个方法:不用全部遍历,而是先从任意节点开始,做bfs,看最后一层节点的个数,如果只有一个,表示从这个节点就可以找到最长的path;如果大于一个,那么从其中任一节点都可以得到最长path, 然后再遍历一次,给每个节点标记层次数,最后找出符合要求的层数数中的节点输出即可!这样一来,复杂度就是o(v+e).
4. 要做两次遍历,还是不方便,不简洁。看了以前的答案,还是用的经典的bfs检查cycle的方法,记录degree,由于这里是undirected graph,所以没有入度和出度一说,直接统计每个节点有几个neighbor。由于已知是没有cycle,那么统计了degree的好处是可以从图的边缘入手去找root,把所有的degree=1的点放在queue,然后每次从queue取出节点,把这些节点的邻居节点的degree减一,如果减一后也碰巧等于1,说明去掉这条边,其邻居也成了新的边界。如此这般,知道n==2或小于2,此时queue里面剩下的就是中心了。
5. 为什么bfs+统计degree可以办到这件事呢?也就是说图的问题很多都可以用bfs+统计degree的放法是为什么?原因,我觉得应该是这种方法可以从结构上结构图,剖析图,把图的性质给显露出来。比如这道题,求最小height,但height又是最长路径,所以从结构入手,就从leaf入手,而degree=1的节点就是leaf,从leaf入手一步一步反推就可以到root
6. 换个思路,我自己想的思路是从root开始做bfs,发现至少都要做两次bfs,有重复之嫌;从相反角度考虑,从leaf入手,就简单一些,而从leaf入手对bfs来说恰好有这个方法:统计入度或度。换个角度看问题而已,无所谓谁好谁坏,提供多角度的原因只是试图还原事物的本来面部,或者试图从多个角度来看事物。如果只看到一个角度,那一定是人偏执的认为只有一个角度,或这某个角度比其他角度好!
//bfs:统计degree
class Solution {
public:
vector<int> findMinHeightTrees(int n, vector<pair<int, int>>& edges) {
//
if(n==1) return {0};
vector<int> degree(n,0);
vector<vector<int>> graph(n);
//step 1:build graph, count the degree
for(auto edge:edges){
graph[edge.first].push_back(edge.second);
graph[edge.second].push_back(edge.first);
degree[edge.first]++;
degree[edge.second]++;
}
queue<int> qq;
//step 2: push the node with 1 degree into queue
for(int i=0;i<n;i++)
if(degree[i]==1) qq.push(i);
//step 3: 利用queue来做bfs,但这个bfs不是从root开始,而是从leaf开始!
while(n>2){
int sz=qq.size();
for(int i=0;i<sz;i++){
int cur=qq.front();
qq.pop();
n--;
for(int j:graph[cur]){
if((--degree[j])==1){
qq.push(j);
}
}
}
}
vector<int> res;
while(!qq.empty()){
res.push_back(qq.front());
qq.pop();
}
/*for(int i=0;i<qq.size();i++)//bug的写法,qq.size()在操作的过程中不能这么干
{
res.push_back(qq.front());
qq.pop();
}*/
return res;
}
};