-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathKNN.m
137 lines (127 loc) · 4.26 KB
/
KNN.m
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
clc;clear;
load usps
k=3;
truerate1=zeros(1,3);
x=1;y=669;
totalsum=0;
distance=zeros(1,7291);
for ii=1:3
true=0;
for i=x:y
for j=1:7291
distance(j)=norm(test(i,:)-train(j,:));%取欧氏距离,得到测试样本与训练样本的欧氏距离
end
[~,train_position]=sort(distance);%排序将欧氏距离从小到大进行排序
train_position=train_position(1,1:k);%取前k个距离在原数据的位置
train_sign=train_number(train_position,1);%取出标签
table=tabulate(train_sign);
[number,Index]=max(table(:,2));%得到频率最高的类别
sign=table(Index,1);
test_sign=test_number(i,1);
if(sign==test_sign)
true=true+1;
end
end
truerate1(1,ii)=true/669;
totalsum=totalsum+truerate1(1,ii);
fprintf('第%d次%d近邻识别usps的正确率为%4.2f\n',ii,k,truerate1(1,ii));
k=k+2;
x=x+669;
y=y+669;
end
fprintf('%d次K近邻识别usps的平均正确率为%4.2f\n',ii,totalsum/ii);
load iris
rbow1=randperm(50);
trainsample_w1=iris1(rbow1(:,1:40),1:5);% 随机取w1类数据中的五分之四，即40组
testsample_w1=iris1(rbow1(:,41:50),1:5);%剩余的10组作为测试样本
rbow1=randperm(50);
trainsample_w2=iris2(rbow1(:,1:40),1:5);%随机取w2类数据中的五分之四，即40组
testsample_w2=iris2(rbow1(:,41:50),1:5);%剩余的10组作为测试样本
rbow1=randperm(50);
trainsample_w3=iris3(rbow1(:,1:40),1:5);%随机取w3类数据中的五分之四，即40组
testsample_w3=iris3(rbow1(:,41:50),1:5);%剩余的10组作为测试样本
trainsample=cat(1,trainsample_w1,trainsample_w2,trainsample_w3); %将三个训练样本合并为一个包含整体的五分之四，即120组数据
testsample=cat(1,testsample_w1,testsample_w2,testsample_w3);%将三个测试样本合并为一个包含整体的五分之一，即30组数据
k=3;
totalsum=0;
truerate2=zeros(1,3);
distance=zeros(1,120);
x=1;y=10;
for ii=1:3
true=0;
for i=x:y
for j=1:120
distance(j)=norm(testsample(i,1:4)-trainsample(j,1:4));
end
[~,train_position]=sort(distance);%排序将欧氏距离从小到大进行排序
train_position=train_position(1,1:k);%取前k个距离在原数据的位置
train_sign=trainsample(train_position,5);%取出标签
table=tabulate(train_sign);
[number,Index]=max(table(:,2));%得到频率最高的类别
sign=table(Index,1);
test_sign=testsample(i,5);
if(test_sign==sign)
true=true+1;
end
end
x=x+10;y=y+10;
truerate2(1,ii)=true/10;
totalsum=totalsum+truerate2(1,ii);
fprintf('第%d次%d近邻识别iris的正确率为%4.2f\n',ii,k,truerate2(1,ii));
k=k+2;
end
fprintf('%d次K近邻识别iris的平均正确率为%4.2f\n',ii,totalsum/ii);
load sonar
data_w1=sonar1; % 随机取w1类数据中的五分之四，即78组
rbow1=randperm(98);
trainsample_w1=data_w1(rbow1(:,1:78),1:61);
testsample_w1=data_w1(rbow1(:,79:98),1:61); %剩余的20组作为测试样本
data_w2=sonar2; %随机取w2类数据中的五分之四，即88组
rbow2=randperm(110);
trainsample_w2=data_w2(rbow2(:,1:88),1:61);
testsample_w2=data_w2(rbow2(:,89:110),1:61);%剩余的22组作为测试样本
trainsample=cat(1,trainsample_w1,trainsample_w2); %将两个训练样本合并为一个作为整体的五分之四，即166组数据
testsample=cat(1,testsample_w1,testsample_w2);%将两个测试样本合并为一个作为包含整体的五分之一，即42组数据
k=3;
totalsum=0;
truerate3=zeros(1,3);
distance=zeros(1,120);
x=1;y=14;
for ii=1:3
true=0;
for i=x:y
for j=1:166
distance(j)=norm(testsample(i,1:60)-trainsample(j,1:60));
end
[~,train_position]=sort(distance);%排序将欧氏距离从小到大进行排序
train_position=train_position(1,1:k);%取前k个距离在原数据的位置
train_sign=trainsample(train_position,61);%取出标签
table=tabulate(train_sign);
[number,Index]=max(table(:,2));%得到频率最高的类别
sign=table(Index,1);
test_sign=testsample(i,61);
if(test_sign==sign)
true=true+1;
end
end
x=x+14;y=y+14;
truerate3(1,ii)=true/14;
totalsum=totalsum+truerate3(1,ii);
fprintf('第%d次%d近邻识别sonar的正确率为%4.2f\n',ii,k,truerate3(1,ii));
k=k+2;
end
fprintf('%d次K近邻识别sonar的平均正确率为%4.2f\n',ii,totalsum/ii);
figure(1)%画图程序
k1=1:3;
plot(truerate1(1,k1),'*')
hold on
plot(truerate2(1,k1),'o')
hold on
plot(truerate3(1,k1),'x')
hold on
legend('Fisher对usps判别','Fisher对iris判别','Fisher对sonar判别');
xlabel('实验次数');
ylabel('准确率%');
grid on;
axis([0 3 0 1]);
title('knn三次分类的准确率');