（3.2）将分词和去停用词后的评论文本基于“环境、卫生、价格、服务”分类

  1 import java.io.BufferedReader;
  2 import java.io.File;
  3 import java.io.FileInputStream;
  4 import java.io.FileNotFoundException;
  5 import java.io.IOException;
  6 import java.io.InputStreamReader;
  7 import java.util.HashSet;
  8 import java.util.Set;
  9 
 10 
 11 public class Fenglei {
 12 
 13     //将分词和去停用词的结果提取出来
 14     String destFile = "." + File.separator + "destFile" + File.separator + "酒店评论.txt";
 15     public static final String RefWord = "." + File.separator + "srcFile" + File.separator + "RefWords.txt";
 16     public static final String EnvironmentRef = "." + File.separator + "srcFile" + File.separator + "EnvironmentRef.txt";
 17     public static final String HealthRef = "." + File.separator + "srcFile" + File.separator + "HealthRef.txt";
 18     public static final String PriceRef = "." + File.separator + "srcFile" + File.separator + "PriceRef.txt";
 19     public static final String ServiceRef = "." + File.separator + "srcFile" + File.separator + "ServiceRef.txt";
 20     
 21     public void fenlei(){
 22         try{
 23             //读取原文件和相关词表
 24             BufferedReader srcBr = new BufferedReader(new InputStreamReader(new FileInputStream(new File(destFile))));
 25             BufferedReader refBr = new BufferedReader(new InputStreamReader(new FileInputStream(new File(RefWord))));
 26             BufferedReader environmentRefBr = new BufferedReader(new InputStreamReader(new FileInputStream(new File(EnvironmentRef))));
 27             BufferedReader healthRefBr = new BufferedReader(new InputStreamReader(new FileInputStream(new File(HealthRef))));
 28             BufferedReader priceRefBr = new BufferedReader(new InputStreamReader(new FileInputStream(new File(PriceRef))));
 29             BufferedReader serviceRefBr = new BufferedReader(new InputStreamReader(new FileInputStream(new File(ServiceRef))));
 30             
 31             //将分词去停用词后的单词，按序放在String数组中
 32             String paragraph = null;
 33             String[] iniArray = null;
 34             for(; (paragraph = srcBr.readLine()) != null;){
 35                 iniArray = paragraph.split(" ");
 36             }            41                         
 42             Set<String> refWordSet = getSet(refBr);        //相关词汇集
 43             Set<String> environmentSet = getSet(environmentRefBr);
 44             Set<String> healthSet = getSet(healthRefBr);
 45             Set<String> priceSet = getSet(priceRefBr);
 46             Set<String> serviceSet = getSet(serviceRefBr);
 47                         
 48             //用一个int数组refNum存放分析文本中寻找相关关键词
 49             int[] refNum = new int[iniArray.length];
 50             int refNumcount = 0;
 51             for(int i = 0; i< iniArray.length; i++){
 52                 if(refWordSet.contains(iniArray[i])){
 53                     refNum[refNumcount++] = i;
 54                 }    
 55             }
 56             
 57             //对refNum数组进行排序
 58             int[] sortRefNum = new int[refNumcount];
 59             bubble(refNum);
 60             for(int i = refNum.length-refNumcount,j=0;j<refNumcount;i++,j++){
 61                 sortRefNum[j] = refNum[i];
 62             }
 63             
 64             //对属于同一相关的短语进行记录
 65             String[] midArray = new String[refNumcount];
 66             int midArrayCount = 0;
 67             for(int i=0; i<refNumcount-1; i++){
 68                 int j=sortRefNum[i];
 69                 midArray[midArrayCount] = "";
 70                 for(;j < sortRefNum[i+1];j++ ){
 71                     if(iniArray[j] != null){                         
 73                         midArray[midArrayCount] += iniArray[j].toString()+" ";                        
 74                     }                    
 75                 }
 76                 midArrayCount++;
 77             }
 78             midArray[midArrayCount] = "";
 79             for(int i = sortRefNum[refNumcount-1];i < iniArray.length;i++){                
 80                 midArray[midArrayCount] += iniArray[i].toString()+" ";
 81             }            
 82             
 83             //对相关短语属于同一大类的短语进行合并整理，如“光线”类和“隔音”应该同属于“环境”类
 84             String[] fourClass = new String[4];
 85             fourClass[0] = "环境： ";
 86             fourClass[1] = "卫生： ";
 87             fourClass[2] = "价格： ";
 88             fourClass[3] = "服务： ";
 89             for(int j=0;j<=midArrayCount;j++){
 90                 if(environmentSet.contains(iniArray[sortRefNum[j]])){   //0-环境
 91                      fourClass[0] += midArray[j].toString()+" ";   
 92                 }else{
 93                     if(healthSet.contains(iniArray[sortRefNum[j]])){    //1-卫生
 94                         fourClass[1] += midArray[j].toString()+" ";
 95                     }else{
 96                         if(priceSet.contains(iniArray[sortRefNum[j]])){  //2-价格
 97                             fourClass[2] += midArray[j].toString()+" ";
 98                         }else{
 99                             if(serviceSet.contains(iniArray[sortRefNum[j]])){   //3-服务
100                                 fourClass[3] += midArray[j].toString()+" ";
101                             }
102                         }
103                     }
104                 }                    
105             }
106         }catch(FileNotFoundException e){
107             // TODO Auto-generated catch block
108             e.printStackTrace();
109         } catch(Exception e){
110             e.printStackTrace();
111         }                
112     }
113     
114     //冒泡排序
115     public int[] bubble(int[] array) {
116         int temp = 0;
117         for (int i = array.length - 1; i > 0; --i) {
118             for (int j = 0; j < i; j++) {
119                 if (array[j] > array[j+1]) {
120                     temp = array[j];
121                     array[j] = array[j+1];
122                     array[j+1] = temp;                
123                 }
124             }
125         }
126         return array;
127     }
128     
129     
130     //词汇表的集合
131     public Set<String> getSet(BufferedReader br){        
132         Set<String> wordSet = new HashSet<String>();    //用来放词汇的集合        
133         String word = null;
134         try {
135             for(; (word = br.readLine()) != null;){
136                 wordSet.add(word);        
137             }
138         } catch (IOException e) {
139             // TODO Auto-generated catch block
140             e.printStackTrace();
141         }                
142         return wordSet;
143     }
144     
145     public static void main(String[] args){
146         new Fenglei().fenlei();
147     }
149 }