本文介绍了如何使用K-means聚类算法对鸢尾花数据集进行聚类分析，通过设置特定的初始中心和聚类数目，对数据集中的150条记录进行分类。分析过程包括数据读取、K-means算法实现、聚类结果输出及最终将分析结果写入Excel文档。
1.按照指定的初始中心、设定的聚类数目实现聚类过程
2.利用下表中数据，初始中心为第1、51、101条数据，聚类数目为3
3.对下表中的150条数据，进行K-means聚类，给出每条记录属于的聚类结果。
表 IRIS数据
|
1-50 |
51-100 |
101-150 |
|
No x1 x2 x3 x4 |
No x1 x2 x3 x4 |
No x1 x2 x3 x4 |
|
1 5.1 3.5 1.4 0.2 |
51 7.0 3.2 4.7 1.4 |
101 6.3 3.3 6.0 2.5 |
|
2 4.9 3.0 1.4 0.2 |
52 6.4 3.2 4.5 1.5 |
102 5.8 2.7 5.1 1.9 |
|
3 4.7 3.2 1.3 0.2 |
53 6.9 3.1 4.9 1.5 |
103 7.1 3.0 5.9 2.1 |
|
4 4.6 3.1 1.5 0.2 |
54 5.5 2.3 4.0 1.3 |
104 6.3 2.9 5.6 1.8 |
|
5 5.0 3.6 1.4 0.2 |
55 6.5 2.8 4.6 1.5 |
105 6.5 3.0 5.8 2.2 |
|
6 5.4 3.9 1.7 0.4 |
56 5.7 2.8 4.5 1.3 |
106 7.6 3.0 6.6 2.1 |
|
7 4.6 3.4 1.4 0.3 |
57 6.3 3.3 4.7 1.6 |
107 4.9 2.5 4.5 1.7 |
|
8 5.0 3.4 1.5 0.2 |
58 4.9 2.4 3.3 1.0 |
108 7.3 2.9 6.3 1.8 |
|
9 4.4 2.9 1.4 0.2 |
59 6.6 2.9 4.6 1.3 |
109 6.7 2.5 5.8 1.8 |
|
10 4.9 3.1 1.5 0.1 |
60 5.2 2.7 3.9 1.4 |
110 7.2 3.6 6.1 2.5 |
|
11 5.4 3.7 1.5 0.2 |
61 5.0 2.0 3.5 1.0 |
111 6.5 3.2 5.1 2.0 |
|
12 4.8 3.4 1.6 0.2 |
62 5.9 3.0 4.2 1.5 |
112 6.4 2.7 5.3 1.9 |
|
13 4.8 3.0 1.4 0.1 |
63 6.0 2.2 4.0 1.0 |
113 6.8 3.0 5.5 2.1 |
|
14 4.3 3.0 1.1 0.1 |
64 6.1 2.9 4.7 1.4 |
114 5.7 2.5 5.0 2.0 |
|
15 5.8 4.0 1.2 0.2 |
65 5.6 2.9 3.9 1.3 |
115 5.8 2.8 5.1 2.4 |
|
16 5.7 4.4 1.5 0.4 |
66 6.7 3.1 4.4 1.4 |
116 6.4 3.2 5.3 2.3 |
|
17 5.4 3.9 1.3 0.4 |
67 5.6 3.0 4.5 1.5 |
117 6.5 3.0 5.5 1.8 |
|
18 5.1 3.5 1.4 0.3 |
68 5.8 2.7 4.1 1.0 |
118 7.7 3.8 6.7 2.2 |
|
19 5.7 3.8 1.7 0.3 |
69 6.2 2.2 4.5 1.5 |
119 7.7 2.6 6.9 2.3 |
|
20 5.1 3.8 1.5 0.3 |
70 5.6 2.5 3.9 1.1 |
120 6.0 2.2 5.0 1.5 |
|
21 5.4 3.4 1.7 0.2 |
71 5.9 3.2 4.8 1.8 |
121 6.9 3.2 5.7 2.3 |
|
22 5.1 3.7 1.5 0.4 |
72 6.1 2.8 4.0 1.3 |
122 5.6 2.8 4.9 2.0 |
|
23 4.6 3.6 1.0 0.2 |
73 6.3 2.5 4.9 1.5 |
123 7.7 2.8 6.7 2.0 |
|
24 5.1 3.3 1.7 0.5 |
74 6.1 2.8 4.7 1.2 |
124 6.3 2.7 4.9 1.8 |
|
25 4.8 3.4 1.9 0.2 |
75 6.4 2.9 4.3 1.3 |
125 6.7 3.3 5.7 2.1 |
|
26 5.0 3.0 1.6 0.2 |
76 6.6 3.0 4.4 1.4 |
126 7.2 3.2 6.0 1.8 |
|
27 5.0 3.4 1.6 0.4 |
77 6.8 2.8 4.8 1.4 |
127 6.2 2.8 4.8 1.8 |
|
28 5.2 3.5 1.5 0.2 |
78 6.7 3.0 5.0 1.7 |
128 6.1 3.0 4.9 1.8 |
|
29 5.2 3.4 1.4 0.2 |
79 6.0 2.9 4.5 1.5 |
129 6.4 2.8 5.6 2.1 |
|
30 4.7 3.2 1.6 0.2 |
80 5.7 2.6 3.5 1.0 |
130 7.2 3.0 5.8 1.6 |
|
31 4.8 3.1 1.6 0.2 |
81 5.5 2.4 3.8 1.1 |
131 7.4 2.8 6.1 1.9 |
|
32 5.4 3.4 1.5 0.4 |
82 5.5 2.4 3.7 1.0 |
132 7.9 3.8 6.4 2.0 |
|
33 5.2 4.1 1.5 0.1 |
83 5.8 2.7 3.9 1.2 |
133 6.4 2.8 5.6 2.2 |
|
34 5.5 4.2 1.4 0.2 |
84 6.0 2.7 5.1 1.6 |
134 6.3 2.8 5.1 1.5 |
|
35 4.9 3.1 1.5 0.2 |
85 5.4 3.0 4.5 1.5 |
135 6.1 2.6 5.6 1.4 |
|
36 5.0 3.2 1.2 0.2 |
86 6.0 3.4 4.5 1.6 |
136 7.7 3.0 6.1 2.3 |
|
37 5.5 3.5 1.3 0.2 |
87 6.7 3.1 4.7 1.5 |
137 6.3 3.4 5.6 2.4 |
|
38 4.9 3.6 1.4 0.1 |
88 6.3 2.3 4.4 1.3 |
138 6.4 3.1 5.5 1.8 |
|
39 4.4 3.0 1.3 0.2 |
89 5.6 3.0 4.1 1.3 |
139 6.0 3.0 4.8 1.8 |
|
40 5.1 3.4 1.5 0.2 |
90 5.5 2.5 5.0 1.3 |
140 6.9 3.1 5.4 2.1 |
|
41 5.0 3.5 1.3 0.3 |
91 5.5 2.6 4.4 1.2 |
141 6.7 3.1 5.6 2.4 |
|
42 4.5 2.3 1.3 0.3 |
92 6.1 3.0 4.6 1.4 |
142 6.9 3.1 5.1 2.3 |
|
43 4.4 3.2 1.3 0.2 |
93 5.8 2.6 4.0 1.2 |
143 5.8 2.7 5.1 1.9 |
|
44 5.0 3.5 1.6 0.6 |
94 5.0 2.3 3.3 1.0 |
144 6.8 3.2 5.9 2.3 |
|
45 5.1 3.8 1.9 0.4 |
95 5.6 2.7 4.2 1.3 |
145 6.7 3.3 5.7 2.5 |
|
46 4.8 3.0 1.4 0.3 |
96 5.7 3.0 4.2 1.2 |
146 6.7 3.0 5.2 2.3 |
|
47 5.1 3.8 1.6 0.2 |
97 5.7 2.9 4.2 1.3 |
147 6.3 2.5 5.0 1.9 |
|
48 4.6 3.2 1.4 0.2 |
98 6.2 2.9 4.3 1.3 |
148 6.5 3.0 5.2 2.0 |
|
49 5.3 3.7 1.5 0.2 |
99 5.1 2.5 3.0 1.1 |
149 6.2 3.4 5.4 2.3 |
|
50 5.0 3.3 1.4 0.2 |
100 5.7 2.8 4.1 1.3 |
150 5.9 3.0 5.1 1.8 |
读取Excel数据，进行K-means算法分析的类：
Window.java
import java.awt.*;
import java.awt.event.*;
import javax.swing.table.DefaultTableModel;
import javax.swing.*;
import java.util.*;
import java.io.*;
import javax.swing.table.*;
import jlfx.UpdateExcel2003;
import java.util.*;
import jxl.*;
import java.lang.Math;
import java.io.File;
public class Window extends JFrame{
JFrame f_major = new JFrame("商务智能");
JTabbedPane tp = new JTabbedPane();
Font ft = new Font("Serif", Font.TRUETYPE_FONT, 18);
Font ft1 = new Font("Serif", Font.ROMAN_BASELINE, 20);
Font ft2 = new Font("Serif", Font.ROMAN_BASELINE, 15);
Font ft3 = new Font("Serif", Font.TRUETYPE_FONT, 16);
JPanel panel = new JPanel();
//控件定义
JLabel yssj_display=new JLabel("------原始数据显示区------", JLabel.CENTER);
JLabel input_l=new JLabel("-------输入-------", JLabel.CENTER);
JLabel jlgs_l=new JLabel("聚类个数", JLabel.CENTER);
JLabel jgxsq_l=new JLabel("-------结果显示区-------", JLabel.CENTER);
JButton yssjshow_b1=new JButton("显示原始数据");
JButton jgshow_b1=new JButton("显示聚类后分析数据");
JButton Cz_b=new JButton("重置");
JTable jTable_1 = new JTable();
JTable jTable_2 = new JTable();
JComboBox num_cb1=new JComboBox();
JScrollPane a1=new JScrollPane(jTable_1,JScrollPane.VERTICAL_SCROLLBAR_ALWAYS,JScrollPane.HORIZONTAL_SCROLLBAR_ALWAYS);
JScrollPane a2=new JScrollPane(jTable_2,JScrollPane.VERTICAL_SCROLLBAR_ALWAYS,JScrollPane.HORIZONTAL_SCROLLBAR_ALWAYS);
class Flower{
int n=0;
double x1=0;
double x2=0;
double x3=0;
double x4=0;
};
Flower[] flower;
int N;
public static void main(String[] args) {
Window win = new Window();
win.go();
}
public void go() {
//窗体界面设计
f_major.setSize(900,600);
f_major.getContentPane().setLayout(new BorderLayout());
f_major.getContentPane().add("Center", tp);
f_major.setFont(ft);
f_major.setVisible(true);
f_major.setResizable(true);
//选项卡界面设计
tp.add("K-means聚类分析", panel);
tp.setFont(ft);
//界面设计
panel.setLayout(null);
//原始数据显示界面
yssj_display.setSize(250, 40);
yssj_display.setLocation(0, 0);
yssj_display.setFont(ft1);
jTable_1.setSize(400, 350);
jTable_1.setLocation(10, 40);
//输入界面
input_l.setSize(185, 40);
input_l.setLocation(0, 390);
input_l.setFont(ft1);
jlgs_l.setSize(100, 40);
jlgs_l.setLocation(10, 430);
jlgs_l.setFont(ft2);
num_cb1.setSize(80, 40);
num_cb1.setLocation(110, 430);
num_cb1.addItem("3");
yssjshow_b1.setSize(120, 30);
yssjshow_b1.setLocation(200,430);
//重置
Cz_b.setSize(150, 40);
Cz_b.setLocation(550, 460);
//结果显示区界面
jgxsq_l.setSize(200, 40);
jgxsq_l.setLocation(450, 0);
jgxsq_l.setFont(ft1);
jTable_2.setSize(400, 350);
jTable_2.setLocation(450, 40);
a1.setViewportView(jTable_1);
a1.setSize(400, 350);
a1.setLocation(10, 40);
a2.setViewportView(jTable_2);
a2.setSize(400, 350);
a2.setLocation(450, 40);
jgshow_b1.setSize(180, 30);
jgshow_b1.setLocation(680, 390);
panel.add(yssj_display);
panel.add(a1);
panel.add(a2);
// panel.add(jTable_1);
panel.add(input_l);
panel.add(jlgs_l);
panel.add(num_cb1);
panel.add(jgxsq_l);
// panel.add(jTable_2);
panel.add(jgshow_b1);
panel.add( yssjshow_b1);
panel.add(Cz_b);
//定义原始数据监控器
yssjshow_b1.addActionListener(new ShowData());
//定义结果显示监控器
jgshow_b1.addActionListener(new ShowSequence());
}
//ShowData内部类
class ShowData implements ActionListener{
public void actionPerformed(ActionEvent arg0) {
try {
InputStream stream = new FileInputStream(new File("jlfx_data.xls"));
Workbook rwb = Workbook.getWorkbook(stream);
Cell cell = null;
Sheet sheet = rwb.getSheet(0);
DefaultTableModel tableModel=(DefaultTableModel) jTable_1.getModel();
tableModel.setColumnCount(5);
tableModel.setRowCount(sheet.getRows());
Object[] object=new Object[jTable_1.getColumnCount()];
N=sheet.getRows()-1;
if(sheet.getRows()>2){
flower=new Flower[sheet.getRows()];
for(int i = 0; i < sheet.getRows(); i++){
flower[i] = new Flower();
}
for(int i=0;i<sheet.getRows();i++){
for(int j=0;j<5;j++)
{
cell=sheet.getCell(j,i);
if(cell.getType()==CellType.LABEL)
{
LabelCell labelcell=(LabelCell)cell;
object[j]=labelcell.getString();
jTable_1.setValueAt(labelcell.getString(),i,j);
}
else if(cell.getType()==CellType.NUMBER)
{
Double numd;
double numi;
NumberCell numc10=(NumberCell)cell;
numd=new Double(numc10.getValue());
numi=numd.doubleValue();
object[j]=numi;
jTable_1.setValueAt(numi,i,j);
if(j==0)flower[i].n=(int)numi;
else if(j==1)flower[i].x1=numi;
else if(j==2)flower[i].x2=numi;
else if(j==3)flower[i].x3=numi;
else if(j==4)flower[i].x4=numi;
}
}
} rwb.close();
}
}
catch (Exception e)
{
System.out.println(e);
}
}
}
//ShowSwquence内部类
class ShowSequence implements ActionListener{
public void actionPerformed(ActionEvent arg0) {
// 获取聚类个数
int k=Integer.parseInt(String.valueOf( num_cb1.getSelectedItem()));
double[][] box=new double[k+1][N+1];
int[][] count ={{0},{0},{0},{0}};
int[][] count_temp ={{0},{0},{0},{0}};
double [][] center = new double[k+1][5];
//初始中心放在一个三行四列的数组里
center[1][1] = flower[1].x1;
center[1][2] = flower[1].x2;
center[1][3] = flower[1].x3;
center[1][4] = flower[1].x4;
center[2][1] = flower[51].x1;
center[2][2] = flower[51].x2;
center[2][3] = flower[51].x3;
center[2][4] = flower[51].x4;
center[3][1] = flower[101].x1;
center[3][2] = flower[101].x2;
center[3][3] = flower[101].x3;
center[3][4] = flower[101].x4;
double[] E={1.0,0,0,0};
double[][] sum_temp={{0,0,0,0,0},{0,0,0,0,0},{0,0,0,0,0},{0,0,0,0,0}};
double x=0.0, temp;
double[] d={0,0,0,0};
int tag=1;
int num=0;
//double[] sum = {0,0,0,0};
while(Math.abs(E[0] - x) > 0.0001)
{
num++;
x = E[0];
for(int i=1; i<4; i++)count[i][0] = 0;
for(int f=1; f<151; f++)
{
//求出所有数据与聚类中心均值的差值存放在d[j]中
for(int j=1; j<4; j++)
{
d[j] = Math.pow(Math.abs(center[j][1] - flower[f].x1), 2);
d[j] += Math.pow(Math.abs(center[j][2] - flower[f].x2), 2);
d[j] += Math.pow(Math.abs(center[j][3] - flower[f].x3), 2);
d[j] += Math.pow(Math.abs(center[j][4] - flower[f].x4), 2);
d[j] = Math.sqrt(d[j]);
}
temp = d[1];
tag=1;
//求出每一行数据与聚类中心均值的欧式距离的最小值，并归于该聚类中心
for(int m=1; m<4; m++)
{
if(d[m] < temp)
{
temp = d[m];
tag = m;//tag标记数据分类到相应的聚类中心
}
}
//每找到一行数据的归属，就将相应的聚类中心的数据行数+1
count[tag][0] = count[tag][0] + 1;
//将某一行数据归于相应的聚类中心的那一行
box[tag][count[tag][0]] = f;
E[tag] += Math.pow(d[tag],2);
//算出每一列的数据的总和
sum_temp[tag][1] += flower[f].x1;
sum_temp[tag][2] += flower[f].x2；
sum_temp[tag][3] += flower[f].x3;
sum_temp[tag][4] += flower[f].x4;
}
E[0] = E[1]+E[2]+E[3];
for(int i=1; i<4; i++)
//3个聚类中心
{
for(int j=1; j<=4; j++)
//4列数据
{
//将每一列数据总和除以每一列的数据的行数，
//得到相应那一列的数据的均值，4列数据的均值便是新的聚类中心
sum_temp[i][j] = sum_temp[i][j]/count[i][0];
//更新了新的聚类中心，赋值给center数组，供下一轮循环使用
center[i][j] = sum_temp[i][j];
sum_temp[i][j] = 0;
}
}
for(int i=1; i<4; i++)
{
E[i] = 0;
d[i] = 0;
count_temp[i][0] = count[i][0];
}
}
System.out.println("循环次数："+num);
//分类完成,输出每个聚类条数；
for(int i=1; i<4; i++)
{
System.out.println("第"+i+"个聚类项数："+count_temp[i][0]);
}
//输出全部聚类内容；
for(int i=1; i<4; i++)
{
for(int j=1; j<=count[i][0]; j++)
{
try{
File file = new File("jieguo.xls");
InputStream stream = new FileInputStream(new File("jieguo.xls")); // 获取Excel文件对象
Workbook rwb = Workbook.getWorkbook(stream);
UpdateExcel.updateExcel(file, 0, (i-1)*4+0, j,flower[(int)box[i][j]].x1);
UpdateExcel.updateExcel(file, 0, (i-1)*4+1, j,flower[(int)box[i][j]].x2);
UpdateExcel.updateExcel(file, 0, (i-1)*4+2, j,flower[(int)box[i][j]].x3);
UpdateExcel.updateExcel(file, 0, (i-1)*4+3, j,flower[(int)box[i][j]].x4);
rwb.close();
}catch (Exception e)
{
System.out.println(e);
}
}
}
try {
InputStream stream = new FileInputStream(new File("jieguo.xls"));
// 获取excel文件对象
Workbook rwb = Workbook.getWorkbook(stream);
Cell cell = null;
Sheet sheet = rwb.getSheet(0);
DefaultTableModel tableModel=(DefaultTableModel) jTable_2.getModel();
tableModel.setColumnCount(sheet.getColumns());
tableModel.setRowCount(sheet.getRows());
Object[] object=new Object[jTable_2.getColumnCount()];
for(int q=0;q<sheet.getRows();q++)
{//列循环
for(int p=0;p<sheet.getColumns();p++)
{
cell=sheet.getCell(p,q);
if(cell.getType()==CellType.LABEL)
{
LabelCell labelcell=(LabelCell)cell;
object[p]=labelcell.getString();
jTable_2.setValueAt(labelcell.getString(),q,p); }
else if(cell.getType()==CellType.NUMBER)
{
Double numd;
double numi;
NumberCell numc10=(NumberCell)cell;
numd=new Double(numc10.getValue());
numi=numd.doubleValue();
object[p]=numi;
jTable_2.setValueAt(numi,q,p);
}
}
}
rwb.close();
}
catch (Exception e)
{
System.out.println(e);
}
}
}
}
将分析结果写入Excel文档的类：
UpDateExcel.java
package jlfx;
import jxl.*;
import jxl.format.UnderlineStyle;
import jxl.write.*;
import jxl.write.Number;
import jxl.write.Boolean;
import org.apache.poi.poifs.filesystem.POIFSFileSystem;
import org.apache.poi.hssf.usermodel.HSSFWorkbook;
import org.apache.poi.hssf.usermodel.HSSFSheet;
import org.apache.poi.hssf.usermodel.HSSFRow;
import org.apache.poi.hssf.usermodel.HSSFCell;
import org.apache.poi.xssf.usermodel.XSSFWorkbook;
import org.apache.poi.xssf.usermodel.XSSFSheet;
import org.apache.poi.xssf.usermodel.XSSFRow;
import org.apache.poi.xssf.usermodel.XSSFCell;
import java.io.*;
public class UpdateExcel {
/**
* @param exlFile
* @param sheetIndex
* @param col
* @param row
* @param value
* @throws Exception
*/
public static void updateExcel(File exlFile, int sheetIndex, int col,
int row, Double value) throws Exception {
FileInputStream fis = new FileInputStream(exlFile);
HSSFWorkbook workbook = new HSSFWorkbook(fis);
// workbook.
HSSFSheet sheet = workbook.getSheetAt(sheetIndex);
HSSFRow r = sheet.getRow(row);
HSSFCell cell = r.createCell(col);
cell.setCellValue(value);
fis.close();
FileOutputStream fos = new FileOutputStream(exlFile);
workbook.write(fos);
fos.close();
}
}
扫一扫
5234
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
