AC自动机 - 多模式串匹配问题的基本运用 + 模板题 --- HDU 2222
Keywords Search
Time Limit: 2000/1000 MS (Java/Others) Memory Limit: 65536/32768 K (Java/Others)
Total Submission(s): 35655 Accepted Submission(s): 11496
Problem Description
In the modern time, Search engine came into the life of everybody like Google, Baidu, etc.
Wiskey also wants to bring this feature to his image retrieval system.
Every image have a long description, when users type some keywords to find the image, the system will match the keywords with description of image and show the image which the most keywords be matched.
To simplify the problem, giving you a description of image, and some keywords, you should tell me how many keywords will be match.
Wiskey also wants to bring this feature to his image retrieval system.
Every image have a long description, when users type some keywords to find the image, the system will match the keywords with description of image and show the image which the most keywords be matched.
To simplify the problem, giving you a description of image, and some keywords, you should tell me how many keywords will be match.
Input
First line will contain one integer means how many cases will follow by.
Each case will contain two integers N means the number of keywords and N keywords follow. (N <= 10000)
Each keyword will only contains characters 'a'-'z', and the length will be not longer than 50.
The last line is the description, and the length will be not longer than 1000000.
Each case will contain two integers N means the number of keywords and N keywords follow. (N <= 10000)
Each keyword will only contains characters 'a'-'z', and the length will be not longer than 50.
The last line is the description, and the length will be not longer than 1000000.
Output
Print how many keywords are contained in the description.
Sample Input
1
5
she
he
say
shr
her
yasherhs
Sample Output
3
Mean:
给你n个单词，再给你一篇文章，让你统计有多少个单词在文章中出现过。
analyse:
裸的AC自动机，模板题。
Time complexity：o(n)+o(ml) n个模式串长度均不超过m,文本串长度为L
Source code：
// Memory Time
// 1347K 0MS
// by : Snarl_jsb
// 2014-09-29-20.14
#include<algorithm>
#include<cstdio>
#include<cstring>
#include<cstdlib>
#include<iostream>
#include<vector>
#include<queue>
#include<stack>
#include<map>
#include<string>
#include<climits>
#include<cmath>
#define N 10010
#define LL long long
using namespace std; namespace ac_auto
{
char str[1000005];
struct node
{
node *next[26];
node *fail;
int count;
node()
{
for(int i = 0; i < 26; i++)
next[i] = NULL;
count = 0;
fail = NULL;
}
}*q[50*N];
node *root;
int head, tail; void Insert(char *str) // 插入单词
{
node *p = root;
int i = 0, index;
while(str[i]) {
index = str[i] - 'a';
if(p->next[index] == NULL)
p->next[index] = new node();
p = p->next[index];
i++;
}
p->count++;
}
void build_ac_automation(node *root) // bfs建立fail指针
{
root->fail = NULL;
q[tail++] = root;
while(head < tail) {
node *temp = q[head++];
node *p = NULL;
for(int i = 0; i < 26; i++) {
if(temp->next[i] != NULL) {
if(temp == root) temp->next[i]->fail = root;
else {
p = temp->fail;
while(p != NULL) {
if(p->next[i] != NULL) {
temp->next[i]->fail = p->next[i];
break;
}
p = p->fail;
}
if(p == NULL) temp->next[i]->fail = root;
}
q[tail++] = temp->next[i];
}
}
}
}
int Query(node *root) // 匹配 + 统计
{
int i = 0, cnt = 0, index;
node *p = root;
while(str[i]) {
index = str[i] - 'a';
while(p->next[index] == NULL && p != root) p = p->fail;
p = p->next[index];
if(p == NULL) p = root;
node *temp = p;
while(temp != root && temp->count != -1) {
cnt += temp->count;
temp->count = -1;
temp = temp->fail;
}
i++;
}
return cnt;
}
}
using namespace ac_auto; int main()
{
int T, n;
scanf("%d",&T);
while(T--)
{
head = tail = 0; // 清零
root = new node(); // 申请新的root结点
scanf("%d",&n);
while(n--)
{
scanf("%s", str);
Insert(str); // 插入单词
}
build_ac_automation(root); // 建树
scanf("%s",str);
printf("%d\n", Query(root)); // 查找+统计
}
return 0;
}
附上注释版(博主很贴心的有木有=.=)
#include<cstdio> const int N = 10010;
char str[1000005];
struct node
{
node *next[26]; // 每个结点都对应26个字母的指针
node *fail; // 失配指针
int count; //
node() // 构造函数初始化
{
for(int i = 0; i < 26; i++)
next[i] = NULL;
count = 0;
fail = NULL;
}
}*q[50*N];
node *root;
int head, tail; void Insert(char *str) // 插入单词.相当于构建一个Trie树
{
node *p = root;
int i = 0, index;
while(str[i]) {
index = str[i] - 'a'; // 转化为相对数字来存
if(p->next[index] == NULL) // 该字母未插入过
p->next[index] = new node(); // 为该字母申请一个结点
p = p->next[index]; // 移至下一个
i++;
}
p->count++; // 记录该结点的单词总共插入的次数
}
void build_ac_automation(node *root) // bfs建立fail指针
{
root->fail = NULL;
q[tail++] = root;
while(head < tail) {
node *temp = q[head++];
node *p = NULL;
for(int i = 0; i < 26; i++) {
if(temp->next[i] != NULL) {
if(temp == root) temp->next[i]->fail = root;
else {
p = temp->fail;
while(p != NULL) {
if(p->next[i] != NULL) {
temp->next[i]->fail = p->next[i];
break;
}
p = p->fail;
}
if(p == NULL) temp->next[i]->fail = root;
}
q[tail++] = temp->next[i];
}
}
}
} int Query(node *root) // 匹配 + 统计
{
int i = 0, cnt = 0, index;
node *p = root;
while(str[i])
{
index = str[i] - 'a';
while(p->next[index] == NULL && p != root) //前缀是相同的，所以不管哪个指针走到了count不为0的结点上，那么该结点所代表的单词就匹配成功
p = p->fail;//失配情况下，p指针指向p->fail.(相当于KMP的next数组)
p = p->next[index];//由于现在所在的位置是父节点，所以需要向下移动一个位置
if(p == NULL)
p = root; //如果匹配失败，移动到root，重新开始匹配
node *temp = p;//
while(temp != root && temp->count != -1) //统计--如果匹配成功,那么count>1，表示该结点代表的单词数量;否则表示该结点没有单词
{
cnt += temp->count; //统计该单词出现的次数
temp->count = -1; //标记为-1，表示该单词已经加入了cnt中，下次就不用重复统计
temp = temp->fail;//判断整条链上的匹配情况
}
i++;
}
return cnt;
} int main()
{
int T, n;
scanf("%d",&T);
while(T--)
{
head = tail = 0; // 清零
root = new node(); // 申请新的root结点
scanf("%d",&n);
while(n--)
{
scanf("%s", str);
Insert(str); // 插入单词
}
build_ac_automation(root); // 建树
scanf("%s",str);
printf("%d\n", Query(root)); // 查找+统计
}
return 0;
}
后来交了一发G++才发现会超内存，毕竟每次Trie树申请的内存没释放，内存开销还是有点大。同样交C++就没事，说明C++的内存回收机制比G++要好很多。
但是自己写了个手动释放内存的函数也还是无济于事。后来又把BFS的静态队列改为了动态队列，同样也没多大作用。
/*
* this code is made by crazyacking
* Verdict: Accepted
* Submission Date: 2015-07-17-13.54
* Time: 0MS
* Memory: 137KB
*/
#include <queue>
#include <cstdio>
#include <set>
#include <string>
#include <stack>
#include <cmath>
#include <climits>
#include <map>
#include <cstdlib>
#include <iostream>
#include <vector>
#include <algorithm>
#include <cstring>
#define LL long long
#define ULL unsigned long long
using namespace std;
const int N = 10010;
const int M = 10 + int( 1e6 );
class Node
{
public:
int cnt;
Node *next[26], *fail;
Node()
{
cnt = 0;
for( int i = 0; i < 26; ++i ) next[i] = NULL;
fail = NULL;
}
};
queue<Node*> q;
//Node *q[50 * N];
Node *root;
int head, tail;
char str[M]; void Insert( char *str ) // build Trie-Tree
{
Node *p = root;
int i = 0, index;
while( str[i] )
{
index = str[i] - 'a';
if( p->next[index] == NULL )
p->next[index] = new Node();
p = p->next[index];
++i;
}
p->cnt++;
} void build_ac_automation( Node *root ) // build fail ptr
{
root->fail = NULL;
while(!q.empty()) q.pop();
q.push(root);
// q[tail++] = root;
while( !q.empty() )
{
Node *temp=q.front();
q.pop();
// Node *temp = q[head++];
Node *p = NULL;
for( int i = 0; i < 26; ++i )
{
if( temp->next[i] != NULL )
{
if( temp == root ) temp->next[i]->fail = root;
else
{
p = temp->fail;
while( p != NULL )
{
if( p->next[i] != NULL )
{
temp->next[i]->fail = p->next[i];
break;
}
p = p->fail;
}
if( p == NULL ) temp->next[i]->fail = root;
}
// q[tail++] = temp->next[i];
q.push(temp->next[i]);
}
}
}
} int Query( Node *root ) // mathing and count
{
int i = 0, ans = 0, index;
Node *p = root;
while( str[i] )
{
index = str[i] - 'a';
while( p->next[index] == NULL && p != root )
p = p->fail;
p = p->next[index];
if( p == NULL )
p = root;
Node *temp = p;
while( temp != root && temp->cnt != -1 )
{
ans += temp->cnt;
temp->cnt = -1;
temp = temp->fail;
}
i++;
}
return ans;
} /**< 开始交G++一直超内存，交C++就过了，看来C++回收内粗的机制做得不错。 */
/**< 手动释放内存也还会超，看来内存占用不在这儿，应该是BFS的静态队列 */
void free_malloc(Node *root)
{
queue<Node*> Q;
Q.push(root);
while(!Q.empty())
{
Node* p=(Q.front());
Q.pop();
for(int i=0;i<26;++i)
{
if(p->next[i]!=NULL)
{
Q.push((Node*) (p->next[i]));
}
}
delete(p);
}
root=NULL;
} int main()
{
ios_base::sync_with_stdio( false );
cin.tie( 0 );
int Cas;
scanf( "%d", &Cas );
while( Cas-- )
{
int n;
head = tail = 0;
root = new Node();
scanf( "%d", &n );
while( n-- )
{
scanf( "%s", str );
Insert( str );
}
build_ac_automation( root );
scanf( "%s", str );
printf( "%d\n", Query( root ) );
free_malloc(root);
}
return 0;
}
/* */
C++类的成员函数会对函数内new申请的空间进行回收，但是写了类依然逃脱不了G++超时的诅咒。
/*
* this code is made by crazyacking
* Verdict: Accepted
* Submission Date: 2015-07-17-16.17
* Time: 0MS
* Memory: 137KB
*/
#include <queue>
#include <cstdio>
#include <set>
#include <string>
#include <stack>
#include <cmath>
#include <climits>
#include <map>
#include <cstdlib>
#include <iostream>
#include <vector>
#include <algorithm>
#include <cstring>
#define LL long long
#define ULL unsigned long long
using namespace std; char str[1000005];
struct node
{
node *next[26];
node *fail;
int count;
node()
{
for( int i = 0; i < 26; i++ )
next[i] = NULL;
count = 0;
fail = NULL;
}
};
queue<node*> q;
node *root; class AC_AUTO_CLASS
{
public:
void Insert( char *str ) // 插入单词
{
node *p = root;
int i = 0, index;
while( str[i] )
{
index = str[i] - 'a';
if( p->next[index] == NULL )
p->next[index] = new node();
p = p->next[index];
i++;
}
p->count++;
}
};
void build_ac_automation( node *root ) // bfs建立fail指针
{
root->fail = NULL;
while( !q.empty() ) q.pop();
q.push( root );
while( !q.empty() )
{
node *temp = q.front(); q.pop();
node *p = NULL;
for( int i = 0; i < 26; i++ )
{
if( temp->next[i] != NULL )
{
if( temp == root ) temp->next[i]->fail = root;
else
{
p = temp->fail;
while( p != NULL )
{
if( p->next[i] != NULL )
{
temp->next[i]->fail = p->next[i];
break;
}
p = p->fail;
}
if( p == NULL ) temp->next[i]->fail = root;
}
q.push( temp->next[i] );
}
}
}
} int Query( node *root ) // 匹配 + 统计
{
int i = 0, cnt = 0, index;
node *p = root;
while( str[i] )
{
index = str[i] - 'a';
while( p->next[index] == NULL && p != root ) p = p->fail;
p = p->next[index];
if( p == NULL ) p = root;
node *temp = p;
while( temp != root && temp->count != -1 )
{
cnt += temp->count;
temp->count = -1;
temp = temp->fail;
}
i++;
}
return cnt;
} int main()
{
int T, n;
scanf( "%d", &T );
while( T-- )
{
root = new node(); // 申请新的root结点
scanf( "%d", &n );
{
AC_AUTO_CLASS ac_auto_class;
while( n-- )
{
scanf( "%s", str );
ac_auto_class.Insert( str ); // 插入单词
}
build_ac_automation( root ); // 建树
scanf( "%s", str );
printf( "%d\n", Query( root ) ); // 查找+统计
}
}
return 0;
}
AC自动机 - 多模式串匹配问题的基本运用 + 模板题 --- HDU 2222的更多相关文章
- AC 自动机——多模式串匹配
网站上的敏感词过滤是怎么实现的呢? 实际上,这些功能最基本的原理就是字符串匹配算法,也就是通过维护一个敏感词的字典,当用户输入一段文字内容后,通过字符串匹配算法来检查用户输入的内容是否包含敏感词. B ...
- AC自动机——多模式串匹配的算法思想
标准KMP算法用于单一模式串的匹配,即在母串中寻求一个模式串的匹配,但是现在又存在这样的一个问题,如果同时给出多个模式串,要求找到这一系列模式串在母串存在的匹配个数,我们应该如何处理呢? 基于KMP算 ...
- AC自动机 - 多模式串的匹配运用 --- HDU 2896
病毒侵袭 Problem's Link:http://acm.hdu.edu.cn/showproblem.php?pid=2896 Mean: 略 analyse: AC自动机的运用,多模式串匹配. ...
- AC自动机 - 多模式串的匹配 --- HDU 3695 Computer Virus on Planet Pandora
Problem's Link Mean: 有n个模式串和一篇文章,统计有多少模式串在文章中出现(正反统计两次). analyse: 好久没写AC自动机了,回顾一下AC自动机的知识. 本题在构造文章的时 ...
- AC自动机 - 多模式串的匹配运用 --- HDU 3065
病毒侵袭持续中 Problem's Link:http://acm.hdu.edu.cn/showproblem.php?pid=3065 Mean: 略 analyse: AC自动机的运用. 这一题 ...
- (17/34)AC自动机/后缀数组/后缀自动机（施工中）
快补题别再摸鱼了(17/34) 1.AC自动机 #define maxnode 1000010 #define maxsize 26 struct ahocT{ int ch[maxnode][max ...
- Aho-Corasick automaton（AC自动机）解析及其在算法竞赛中的典型应用举例
摘要: 本文主要讲述了AC自动机的基本思想和实现原理,如何构造AC自动机,着重讲解AC自动机在算法竞赛中的一些典型应用. 什么是AC自动机? 如何构造一个AC自动机? AC自动机在算法竞赛中的典型应用 ...
- HDU - 2222,HDU - 2896,HDU - 3065,ZOJ - 3430 AC自动机求文本串和模式串信息(模板题)
最近正在学AC自动机,按照惯例需要刷一套kuangbin的AC自动机专题巩固 在网上看过很多模板,感觉kuangbin大神的模板最为简洁,于是就选择了用kuangbin大神的模板. AC自动机其实就是 ...
- 字符串（3）AC自动机
AC自动机真神奇,其实说白了就是在trie树上进行kmp模式匹配,不过刚接触确实有些难度,有些思想确实有些难以理解,所以学习的时候最好亲自手动模拟整个算法的全过程,那我就来写篇blog总结一下. 首先 ...
随机推荐
- 高大上技术之sql解析
Question: 为何sql解析和高大上有关系?Answer:因为数据库永远都是系统的核心,CRUD如此深入码农的内心...如果能把CRUD改造成高大上技术,如此不是造福嘛... CRUD就是Cre ...
- 每日英语：why can't China produce world-class CEO?
The appointment of India-born Satya Nadella as Microsoft Corp.'s CEO has caused a bit of a stir in C ...
- 【Theano】安装Theano
http://blog.csdn.net/niuwei22007/article/details/47684673 http://blog.csdn.net/m624197265/article/de ...
- zxing扫码--镭射线
同步发表于http://avenwu.net/2015/09/15/zxing_view_finder_laser 在很多应用中都有二维码扫描的需求,比如微信,通过扫描电脑二维码,实现用户登录授权: ...
- Java 中类的加载顺序
如果类A和类B中有静态变量,静态语句块,非静态变量,非静态语句块,构造函数,静态方法,非静态方法,同时类A继承类B,请问当实例化A时,类内部的加载顺序是什么? 测试代码如下: Class B: pub ...
- GPUImage 内置滤镜解析
#pragmamark - 调整颜色 Handle Color GPUImageBrightnessFilter //亮度GPUImageExposureFilter //曝光GPUImageCont ...
- C#标准响应数据
public HttpResponseMessage UpdateModule(Mode mode) { var response = Process.Instance.ExecuteString(( ...
- 日暮·第一章·决斗
日暮 第一章 决斗 泉州府,位于帝国的东南沿海,在数百年前,这里已是帝国最大的通商口岸之一,其一城之繁荣喧哗足以与异邦小国的都城相媲美,无数的人曾经来到这里,追逐财富,梦想,女人以及所有他们认为可 ...
- NPOIHelper.cs (NPOI 2.1.1)
using System; using System.Collections.Generic; using System.Linq; using System.Text; using System.D ...
- linux上java路径设置
linux上java路径设置 标签: javalinuxpath 2012-04-30 22:14 1843人阅读 评论(0) 收藏 举报 版权声明:本文为博主原创文章,未经博主允许不得转载. 往/e ...