《基于朴素贝叶斯分类器的文本分类算法(C语言).doc》由会员分享，可在线阅读，更多相关《基于朴素贝叶斯分类器的文本分类算法(C语言).doc（7页珍藏版）》请在taowenge.com淘文阁网|工程机械CAD图纸|机械工程制图|CAD装配图下载|SolidWorks_CaTia_CAD_UG_PROE_设计图分享下载上搜索。
1、基于朴素贝叶斯分类器的文本分类算法(C语言).txt两个人吵架，先说对不起的人，并不是认输了，并不是原谅了。他只是比对方更珍惜这份感情。#include #include #include /_getcwd(), _chdir()#include /_MAX_PATH, system()#include /_finddata_t, _findfirst(), _findnext(), _findclose()char vocabulary100020;/*声明公有二维数组，用来存储分割好的单词*/*=将要分类的文本分割成单词存储在二维数组vocabulary中=*/输入参数：要分类的文本/输出
2、参数：该文本中总单词数int SplitToWord(char text)int i=0;char seps=, .n; /*定义单词的分隔符*/ char *substring; /*利用分隔符将文本内容分割成单词并存储*/substring=strtok(text,seps); while(substring!=NULL) strcpy(vocabularyi,substring);/将单词存储到vocabulary数组中 substring=strtok(NULL,seps); i+;return i; /返回一共多少个单词/*=计算该目录下的文件数=*/输入参数：无/输出参数：该目录下
3、.txt文件数int CountDirectory()int count=0; /txt文件计数器long hFile; _finddata_t fileinfo;/*查找.txt文件，记录文件数*/ if (hFile=_findfirst(*.txt,&fileinfo)!=-1L) do count+; while (_findnext(hFile,&fileinfo) = 0);return count;/*=计算某类别中P(ai|vj)=*/输入参数：分类文本中单词数/输出参数：该类别下P(ai|vj)float CalculateWordProbability(int wordCo
4、unt)int countSame; /分类文本中的某单词在所有训练样本中出现次数int countAll=0; /训练样本中总单词数char token;FILE *fp;float wordProbability=1; /为后面联乘做准备int i,j;long hFile; _finddata_t fileinfo;for(j=0;jwordCount;j+) /对于分类样本中的每一个单词 countSame=0; countAll=0; if(hFile=_findfirst(*.txt,&fileinfo)!=-1L) /对于该类别下每一个.txt文本 do if(fp=fopen(
5、fileinfo.name,r)=NULL) /是否能打开该文本 printf(Sorry!Cannot open the file!n); exit(0); /*存储此.txt文件中每个单词并与分类文本的单词作比较*/ while(token = fgetc(fp) != EOF) char keyword1024; i = 0; keyword0 = token; / 将每个词第一个字符赋给数组第一个元素 while (keyword+i = fgetc(fp) != & keywordi != t & keywordi != EOF & keywordi != n); / 开始读字符，直
6、到遇到空白符，说明找到一个词 keywordi = 0;/ 加结束符 countAll+; if (strcmp(keyword,vocabularyj) = 0) /比较两个单词是否相同 countSame+; fclose(fp); while (_findnext(hFile,&fileinfo) = 0); wordProbability*=(float)(countSame+1)/(float)(wordCount+countAll)*300; /计算P(wj|vi)，为了扩大效果而*380return wordProbability; /*=计算每个类别的最终概率输出结果=*/输入
7、参数：分类文本中单词数 void CalculateProbability(int wordCount)/*将类别表存储在二维数组中*/FILE *fp;char classList1020; /类别列表 char ch; /临时读取字符使用 int index=0; /classList的行标 int className_c=0; /classList的列标if(fp=fopen(ClassList.txt,r)=NULL) printf(Failed to open the file: ClassList.txt.n); ch = fgetc(fp); while(ch!=EOF) if(
8、ch!=n) classListindexclassName_c=ch; className_c+; else classListindexclassName_c=0; index+; className_c=0; ch = fgetc(fp);/*计算总文本数和每个类别下的文本数、P(ai|vj)*/int txtCount10; /每个类别下的训练文本数int countAll=0; /训练集中总文本数float wordProbability10; /每个类别的单词概率，即P(ai|vj)if(_chdir(F:SogouCSample1) /更改当前绝对路径 printf(系统找不到指
9、定路径!n);else txtCount0=CountDirectory(); /获取该类别下.txt文件数 countAll+=txtCount0; wordProbability0=CalculateWordProbability(wordCount); /获取该类别下P(wj|vi)if(_chdir(F:SogouCSample2) /更改当前绝对路径 printf(系统找不到指定路径!n);else txtCount1=CountDirectory(); /获取该类别下.txt文件数 countAll+=txtCount1; wordProbability1=CalculateWor
10、dProbability(wordCount); /获取该类别下P(wj|vi)if(_chdir(F:SogouCSample3) /更改当前绝对路径 printf(系统找不到指定路径!n);else txtCount2=CountDirectory(); /获取该类别下.txt文件数 countAll+=txtCount2; wordProbability2=CalculateWordProbability(wordCount); /获取该类别下P(wj|vi)if(_chdir(F:SogouCSample4) /更改当前绝对路径 printf(系统找不到指定路径!n);else txt
11、Count3=CountDirectory(); /获取该类别下.txt文件数 countAll+=txtCount3; wordProbability3=CalculateWordProbability(wordCount); /获取该类别下P(wj|vi)if(_chdir(F:SogouCSample5) /更改当前绝对路径 printf(系统找不到指定路径!n);else txtCount4=CountDirectory(); /获取该类别下.txt文件数 countAll+=txtCount4; wordProbability4=CalculateWordProbability(wo
12、rdCount); /获取该类别下P(wj|vi)if(_chdir(F:SogouCSample6) /更改当前绝对路径 printf(系统找不到指定路径!n);else txtCount5=CountDirectory(); /获取该类别下.txt文件数 countAll+=txtCount5; wordProbability5=CalculateWordProbability(wordCount); /获取该类别下P(wj|vi)if(_chdir(F:SogouCSample7) /更改当前绝对路径 printf(系统找不到指定路径!n);else txtCount6=CountDir
13、ectory(); /获取该类别下.txt文件数 countAll+=txtCount6; wordProbability6=CalculateWordProbability(wordCount); /获取该类别下P(wj|vi)if(_chdir(F:SogouCSample8) /更改当前绝对路径 printf(系统找不到指定路径!n);else txtCount7=CountDirectory(); /获取该类别下.txt文件数 countAll+=txtCount7; wordProbability7=CalculateWordProbability(wordCount); /获取该类
14、别下P(wj|vi)if(_chdir(F:SogouCSample9) /更改当前绝对路径 printf(系统找不到指定路径!n);else txtCount8=CountDirectory(); /获取该类别下.txt文件数 countAll+=txtCount8; wordProbability8=CalculateWordProbability(wordCount); /获取该类别下P(wj|vi)if(_chdir(F:SogouCSample10) /更改当前绝对路径 printf(系统找不到指定路径!n);else txtCount9=CountDirectory(); /获取该
15、类别下.txt文件数 countAll+=txtCount9; wordProbability9=CalculateWordProbability(wordCount); /获取该类别下P(wj|vi)/*计算先验概率和最终概率并输出分类结果*/float max=0;int classNo=0;float priorProbability10;float finalProbability10;for(int i=0;imax) /找到最大概率并记录 max=finalProbabilityi; classNo=i; printf(该文本为类别%s的概率为:%.5en,classListi,f
16、inalProbabilityi); /输出每个类别的最终概率printf(n经分析，该文本最有可能为%s类文本!n,classListclassNo); /输出最后分类结果/*=调用文本分割函数和计算最终概率函数=*/输入参数：分类文本void NaiveBayesClassifier(char text)int vocabularyCount;/分类样本中单词数vocabularyCount=SplitToWord(text); /对要分类的文本进行单词分割，结果存储在vocabulary数组中，返回分类样本中单词数CalculateProbability(vocabularyCount)
17、; /计算最终概率/*=程序入口=*/int main()char text=Microsoft offered 44.6 billion dollars to buy Yahoo.February 1st network reported the Associated Press news, Microsoft offered 44.6 billion dollars in cash and stock to buy Yahoo search site.Microsoft offered to pay 31 dollars per share for Yahoo.Microsofts acq
18、uisition offer on Jan. 31 premium of 62% than Yahoos closing price of 19.18 dollars.Microsoft said that Yahoo shareholders can choose cash or stock transactions. Microsoft and Yahoo have sought cooperation in late 2006 and early 2007.The last two years, Yahoo has been in a dilemma: the market share
19、decline,poor operating performance,stock prices tumbled sharply.Trying to make a difference for Microsoft in the Internet market, the acquisition of Yahoo is a shortcut, because the two sides have very strong complementarity.;/*要分类的文本内容*/ NaiveBayesClassifier(text); /*调用朴素贝叶斯分类函数，返回最终分类结果*/return 1;