标签:
Given a singly linked list, return a random node‘s value from the linked list. Each node must have the same probability of being chosen.
Follow up:
What if the linked list is extremely large and its length is unknown to you? Could you solve this efficiently without using extra space?
Example:
1 // Init a singly linked list [1,2,3].
2 ListNode head = new ListNode(1);
3 head.next = new ListNode(2);
4 head.next.next = new ListNode(3);
5 Solution solution = new Solution(head);
6
7 // getRandom() should return either 1, 2, or 3 randomly. Each element should have equal probability of returning.
8 solution.getRandom();
思路:从n个数中选k个数,每个数被选中的几率是k/n的方法:
将n个数的前k个数选出来,构成一个备选集合。然后从第k+1个数开始向后遍历直到第n个数。遍历第k+1个数时,我们用k/(k+1)的几率选中它,然后随机替换掉备选集合中的一个数。则此时对于备选集合中的每个数,不被替换掉的几率是 1 - P(第k+1个数被选中并替换掉该数) = 1 - k/(k+1) * 1/k = k/(k+1)。
假设遍历到第i个数时,一个数留在备选集合中的概率是k/i。则对于第i+1个数,我们用k/(i+1)的概率选中它,并随机替换掉备选集合中的一个数,则对于备选集合中的每一个数,仍然留在集合中的概率是P(该数在上一次流了下来)(1-P(第i+1个数被选中并替换掉该数))= k/i * (1 - k/(i+1) * 1/k) = k/(i+1)。
因此,当我们进行到第n个数时,该数进入备选集合的概率是k/n, 备选集合中其他的数留下的概率是k/n。最后备选集合中的数就是我们要随机选出的k个数,每个数被选出的几率是k/n。
这个题里,让k等于1即可。因此备选集合中始终只有一个值,遍历第i个数时,用1/i的概率选中它并与备选值进行替换。最后的备选值就是随机选出的数,选取的概率为1/n。
1 /** 2 * Definition for singly-linked list. 3 * struct ListNode { 4 * int val; 5 * ListNode *next; 6 * ListNode(int x) : val(x), next(NULL) {} 7 * }; 8 */ 9 class Solution { 10 public: 11 ListNode* head; 12 /** @param head The linked list‘s head. 13 Note that the head is guaranteed to be not null, so it contains at least one node. */ 14 Solution(ListNode* head) { 15 this->head = head; 16 } 17 18 /** Returns a random node‘s value. */ 19 int getRandom() { 20 ListNode* res; 21 int count = 1; 22 for (ListNode* i = head; i != NULL; i = i->next, count++) { 23 srand(time(NULL)); 24 int roll = rand() % count + 1; 25 if (roll == 1) res = i; 26 } 27 return res->val; 28 } 29 }; 30 31 /** 32 * Your Solution object will be instantiated and called as such: 33 * Solution obj = new Solution(head); 34 * int param_1 = obj.getRandom(); 35 */
Linked List Random Node -- LeetCode
标签:
原文地址:http://www.cnblogs.com/fenshen371/p/5798837.html