java xml 大文件怎么打开_JAVA xml 流方式读取。数据挖掘大文件预处理。

importjava.io.BufferedReader;importjava.io.BufferedWriter;importjava.io.File;importjava.io.FileReader;importjava.io.FileWriter;importjava.io.IOException;importjava.util.HashMap;importjava.util.Iterator;importjava.util.Map;/***@authorgjf

*db_pre.arff存储的是从xml文件中读取的xml信息*/

public classElmAuth {

Map map = new HashMap();//第一步//从xml文件中提取 读取xml文件中的author信息，写到db_pre.arff，并且替换特殊字符

public void settleXml(String src, String dst){//src=dblp.xml dst=db_pre.arff

File file = newFile(src);

File fl= newFile(dst);

FileReader fr;try{

fr= newFileReader(file);

FileWriter fw= newFileWriter(fl);

BufferedReader br= newBufferedReader(fr);

BufferedWriter bw= newBufferedWriter(fw);

String line= null;boolean flag = true;intloc_st;intloc_end;int len = 0, max = 0;while((line = br.readLine()) != null){if(line == null)break;

loc_st= line.indexOf("");if(loc_st != -1){

loc_end= line.indexOf("");

line= line.substring(loc_st + 8, loc_end);//在之间的数据,一个作者的名字

line=line.replace('&', ' ');

line=line.replace('$', ' ');

line=line.replace("' "," ");

line=line.replace("'", " ");/*flag以文章为界限,在同一篇文章内,flag=false,写入在同一行*/

if(flag){

bw.write("\n");

bw.write(line);

}else{

bw.write(",");

bw.write(line);

}

len++;//每写一个作者,计数加 +1

flag = false;

}else{

flag= true;if(max < len) max = len;//选择最大的len;

len = 0;

bw.flush();

}

}

System.out.println("第一步 论文中具有最大的作者数:" +max);

}catch(IOException e) {

e.printStackTrace();

}

}//消除只有单个作item//第二步:将作者的信息db_pre.arff中只有一个作者的数据删除

public void elimate_one(String src, String dst){//src=db_pre.arff dst=db_elone.arff

try{

File file= newFile(src);

FileReader fr= newFileReader(file);

BufferedReader br= newBufferedReader(fr);

File filew= newFile(dst);

FileWriter fw= newFileWriter(filew);

BufferedWriter bw= newBufferedWriter(fw);

Map map = new HashMap();

String line= null;int k = 1;int res = 0;while((line = br.readLine()) != null){

String[] arrLine= line.split(",");//作者之间用","隔离,","的数量表示作者的个数,数量比一少,则不写入.

if(arrLine.length > 1){

bw.write(line);

bw.write("\n");

res++;

}

}

bw.flush();

br.close();

bw.close();

fr.close();//System.out.println("The Row of the file is:" + res);

System.out.println("这篇论文中去除单个作者后的行数:" +res);

}catch(IOException e) {

e.printStackTrace();

}

}//将剩余的作储再hashMap中，key值为人名，value为出现的次数,支持度数

public void createMap(String src){//srr=db_elone.arff

try{

File file= newFile(src);

FileReader fr= newFileReader(file);

BufferedReader br= newBufferedReader(fr);

String line= null;while((line = br.readLine()) != null){if(line == null)break;

String[] arrLine= line.split(",");for(int i = 0; i < arrLine.length; ++i){if(map.get(arrLine[i]) == null){

map.put(arrLine[i],1);

}else{

map.put(arrLine[i], map.get(arrLine[i])+ 1);

}

}

}

fr.close();

br.close();

}catch(IOException e) {

e.printStackTrace();

}

}//从hashMap中删除小于支持度minsup的作者,本次的支持度数为100;

public void settleMap(intminsup){

Iterator it=map.keySet().iterator();while(it.hasNext()){

String str=(String) it.next();if(map.get(str)

it.remove();

}

}

System.out.println("Map的大小,支持度大于100的作者个数:" +map.size());

}//将大于minsup的作者存储到文件 db_minsup.arff,存储的是符合筛选的作者

public void updateMap(String src, String dst){//src=db_elone.arff dst=db_minsup.arff

try{

File filer= newFile(src);

FileReader fr= newFileReader(filer);

BufferedReader br= newBufferedReader(fr);

File filew= newFile(dst);

FileWriter fw= newFileWriter(filew);

BufferedWriter bw= newBufferedWriter(fw);

String line= null;int res = 0;boolean flag = true;while((line = br.readLine()) != null){if(line == null)break;

String[] arrLine= line.split(",");if(flag == false)res++;

flag= true;for(int i = 0; i < arrLine.length; ++i){if(map.get(arrLine[i]) != null){if(flag == true){

bw.write("\n" +arrLine[i]);

flag= false;

}else{

bw.write("," +arrLine[i]);

}

}

}

}

bw.flush();

System.out.println("符合筛选的作者合作写的论文篇数:" +res);

fw.close();

bw.close();

fr.close();

br.close();

}catch(IOException e) {

e.printStackTrace();

}

}//生成weka识别的文 dst=db

public void createWekaFile(String src, String dst){//src=db_minsup.arff dst=db

try{

File filer= newFile(src);

FileReader fr= newFileReader(filer);

BufferedReader br= newBufferedReader(fr);

File filew= newFile(dst);

FileWriter fw= newFileWriter(filew);

BufferedWriter bw= newBufferedWriter(fw);

bw.write("@relation db" + "\n");

Iterator it=map.keySet().iterator();while(it.hasNext()){

String str=(String) it.next();

str.replace("'", "\'");

bw.write("@attribute '" + str + "' { t}\n");

}

bw.write("@data" + "\n");

String line= null;boolean flag = true;while((line = br.readLine()) != null){if(line == null)break;

flag= true;charch;

it=map.keySet().iterator();while(it.hasNext()){

String str=(String)it.next();if(line.indexOf(str) >= 0){

ch= 't';

}else{

ch= '?';

}if(flag == true){

bw.write(ch);

}else{

bw.write("," +ch);

}

flag= false;

}

bw.write("\n");

}

bw.flush();

fw.close();

bw.close();

fr.close();

br.close();

}catch(IOException e) {//TODO Auto-generated catch block

e.printStackTrace();

}

}public voidclearMap(){

map.clear();

}public static voidmain(String args[]){

ElmAuth elmauth= newElmAuth();

elmauth.settleXml("dblp.xml", "db_pre.arff");

elmauth.elimate_one("db_pre.arff", "db_elone.arff");

elmauth.createMap("db_elone.arff");

elmauth.settleMap(100);//确定最小支持度数

elmauth.updateMap("db_elone.arff", "db_minsup.arff");for(int i = 0; i < 20; ++i){

System.out.println();

elmauth.elimate_one("db_minsup.arff", "db_minsup_elone.arff");

elmauth.clearMap();

elmauth.createMap("db_minsup_elone.arff");

elmauth.settleMap(100);

elmauth.updateMap("db_minsup_elone.arff", "db_minsup.arff");

}

elmauth.createWekaFile("db_minsup.arff", "db.arff");

}

}