요약 ¶
- 가장 ëŠ�ë¦¬ê³ ë¬´ì‹�한 Linear Search로ë�„ ë¬¸ì œí•´ê²° 했다를 보여주는 ì�˜ì§€ì�˜ 한êµì�¸ 코드
- 프로그램ì�„ 재실행하면 ë˜� 오래 ê¸°ë‹¤ë ¤ì•¼ 해서 파ì�¼ ìž…ì¶œë ¥ì�„ ì �ê·¹ 활용
- ê³ ë¡œ ì�´ 프로그램ì�„ 복붙해서 그냥 실행하면 ë�Œì•„가지 않습니다
- ì�´ 프로그램ì�€ Bayes Classifier ê°’ì�„ 구하는것 까지ì�´ê³ , 시간 관계ìƒ� ê°’ì�˜ ì°¸/ê±°ì§“ 빈ë�„는 ì—‘ì…€ì�„ ì�´ìš©í•´ì„œ 계산했습니다.
í™•ë¥ ê°’ ë¶„ì„� ¶
- 다른 분들과 달리, ì €ëŠ” 한 글ìž�ì�¸ 문ìž�와 특수문ìž�를 첫 글ìž�로 í�¬í•¨í•˜ëŠ” 단어들ì�€ Train Data ë°� Test Data ë¶„ì„�ì—� í�¬í•¨ì‹œí‚¤ì§€ 않았습니다.
- 때문ì—�, ê²½ì œ ë©´ì—�서는 200ê°œì�˜ 기사 중 148ê°œì�˜ 분류를 성공하는 ì €ì¡°í•œ 성ì �ì�„ 보였습니다.
- 반면, ì •ì¹˜ ë©´ì—�서는 200ê°œì�˜ 기사 중 188ê°œì�˜ 분류를 성공하여, 94%ì�˜ ì •í™•ë�„를 보였습니다.
Source Code ¶
- 여러분ì�˜ ì •ì‹ ê±´ê°•ì�„ 위해.. ì�´ 코드를 ë¶„ì„�하는 것ì�„ 권장하지는 않습니다;ã…�;
Train File Analysis ¶
import java.io.*;
import java.util.ArrayList;
import java.util.List;
import java.util.StringTokenizer;
public class FileAnalasys {
String filename;
List<String> articles;
public List<String> data;
public List<Integer> frequency;
FileAnalasys(String f){
filename = f;
articles = new ArrayList<String>();
data = new ArrayList<String>();
frequency = new ArrayList<Integer>();
}
public void moveFileToData(){
try{
FileReader fr = new FileReader(filename);
BufferedReader br = new BufferedReader(fr);
String line;
String str;
line = br.readLine();
while(line != null){
articles.add(line);
StringTokenizer st = new StringTokenizer(line, " ");
while(st.hasMoreTokens()){
str = st.nextToken();
if(str.length() > 1 && str.charAt(0) != '.' && str.charAt(0) != ','&& str.charAt(0) != '!'&& str.charAt(0) != '"' && str.charAt(0) != ':' && str.charAt(0) != '-' && str.charAt(0) != ';'){
boolean addFlag = true;
for(int i=0; i<data.size(); i++){
if(str.equals(data.get(i))){
frequency.set(i, frequency.get(i)+1);
addFlag = false;
break;
}
}
if(addFlag){
data.add(str);
frequency.add(1);
}
}
}
line = br.readLine();
}
this.printDataToFile();
fr.close();
} catch(IOException e){
e.printStackTrace();
}
}
private void printDataToFile(){
try {
String fname = "D:/analysis.txt";
FileWriter fw = new FileWriter(fname);
String str = null;
for(int i=0; i < data.size(); i++){
str = data.get(i) + "\t" + frequency.get(i) + "\n";
fw.write(str);
}
fw.close();
} catch (IOException e) {
e.printStackTrace();
}
}
public void loadAnalysisFile(){
try{
FileReader fr = new FileReader(filename);
BufferedReader br = new BufferedReader(fr);
String line;
line = br.readLine();
while(line != null){
StringTokenizer st = new StringTokenizer(line, "\t");
data.add(st.nextToken());
frequency.add(Integer.parseInt(st.nextToken()));
line = br.readLine();
}
} catch(IOException e){
e.printStackTrace();
}
}
public void printArticle(){
for(int i=0; i<articles.size(); i++){
System.out.println(articles.get(i));
System.out.print("\n\n");
}
}
public void printData(){
for(int i=0; i<data.size(); i++){
System.out.print(data.get(i));
System.out.println(" " + frequency.get(i));
}
System.out.println("Data size : "+data.size());
System.out.println("Frequency size : "+frequency.size());
}
public int getArticleLength(){ return articles.size(); }
}
test File Analysis ¶
import java.io.BufferedReader;
import java.io.FileReader;
import java.io.FileWriter;
import java.io.IOException;
import java.util.ArrayList;
import java.util.List;
import java.util.StringTokenizer;
public class testFileCal {
String filename;
List<Double> bayes;
testFileCal(String f){
filename = f;
bayes = new ArrayList<Double>();
}
public void testArticleRead(List<String> data, List<Integer> frequency, List<String> data2, List<Integer> frequency2){
try{
FileReader fr = new FileReader(filename);
BufferedReader br = new BufferedReader(fr);
String line;
String str;
line = br.readLine();
while(line != null){
StringTokenizer st = new StringTokenizer(line, " ");
ArrayList<String> articleWords = new ArrayList<String>();
while(st.hasMoreTokens()){
str = st.nextToken();
if(str.length() > 1 && str.charAt(0) != '.' && str.charAt(0) != ','&& str.charAt(0) != '!'&& str.charAt(0) != '"' && str.charAt(0) != ':' && str.charAt(0) != '-' && str.charAt(0) != ';'){
boolean addFlag = true;
for(int j=0; j<articleWords.size(); j++){
if(str.equals(articleWords.get(j))){
addFlag = false;
break;
}
}
if(addFlag) articleWords.add(str);
}
}
for(int i=0; i<articleWords.size(); i++) System.out.println(articleWords.get(i));
double bayesNumber = 0;
for(int j=0; j<articleWords.size(); j++){
int eNum = 0;
int pNum = 0;
for(int k=0; k<data.size(); k++){
if(articleWords.get(j).equals(data.get(k))) eNum = frequency.get(k);
}
for(int k=0; k<data2.size(); k++){
if(articleWords.get(j).equals(data2.get(k))) pNum = frequency2.get(k);
}
if (eNum == 0 || pNum == 0){ eNum++; pNum++; }
bayesNumber += Math.log((double)eNum / (double)pNum);
System.out.println(bayesNumber);
}
bayes.add(bayesNumber);
line = br.readLine();
}
} catch (IOException e) {
e.printStackTrace();
}
}
public void printBayes(){
System.out.println("bayes size : "+bayes.size());
try {
String fname = "D:/bayesResult.txt";
FileWriter fw = new FileWriter(fname);
String str = null;
for(int i=0; i < bayes.size(); i++){
str = bayes.get(i) + "\n";
System.out.println(str);
fw.write(str);
}
fw.close();
} catch (IOException e) {
e.printStackTrace();
}
}
public static void main(String [] args){
FileAnalasys ec = new FileAnalasys("D:/economy_analysis.txt");
FileAnalasys po = new FileAnalasys("D:/politics_analysis.txt");
ec.loadAnalysisFile();
po.loadAnalysisFile();
//testFileCal testE = new testFileCal("D:/economy.txt");
testFileCal testP = new testFileCal("D:/politics.txt");
//testE.testArticleRead(ec.data, ec.frequency, po.data, po.frequency);
testP.testArticleRead(po.data, po.frequency, ec.data, ec.frequency);
testP.printBayes();
}
}










