Teach with us

How to find the confusion matrix for linear discriminant analysis in R?

R ProgrammingServer Side ProgrammingProgramming

To find the confusion matrix for linear discriminant analysis in R, we can follow the below steps −

  • First of all, create a data frame.
  • Create new features using linear discriminant analysis.
  • Find the confusion matrix for linear discriminant analysis using table and predict function.

Create the data frame

Let's create a data frame as shown below −

 Live Demo

Group<-sample(c("I","II","III","IV"),25,replace=TRUE)
Score1<-sample(1:10,25,replace=TRUE)
Score2<-sample(1:10,25,replace=TRUE)
Score3<-sample(1:10,25,replace=TRUE)
Score4<-sample(1:10,25,replace=TRUE)
df<-data.frame(Group,Score1,Score2,Score3,Score4)
df

On executing, the above script generates the below output(this output will vary on your system due to randomization) −

   Group Score1 Score2 Score3  Score4
1    IV     7      5       2     5
2    III    5      3       2     4
3    III    8      9       4     7
4    IV     6      1       1     5
5    III    8      4       6     8
6    IV     9      2       1     7
7    I      3      2       3     2
8    IV     5      8       3     3
9    II     7      4       4     1
10   IV     5      4       1     10
11   II     3      1       2     4
12   III    3      2       1     7
13   IV     1      4       7     6
14   III    10     8       9     2
15   II     3      7       8     1
16   I      9      2       3     1
17   III    2      7       3     2
18   IV     7      7       1     7
19   IV     2      6       1     3
20   I      4      10      6     1
21   I      1      6       4     4
22   I      6      3       6     2
23  III     6      6       3     5
24   I      2      3       10    10
25   II     4      4       2      5

Create new features using linear discriminant analysis

Use lda function of MASS package to find the new features for data in data frame df −

 Live Demo

Group<-sample(c("I","II","III","IV"),25,replace=TRUE)
Score1<-sample(1:10,25,replace=TRUE)
Score2<-sample(1:10,25,replace=TRUE)
Score3<-sample(1:10,25,replace=TRUE)
Score4<-sample(1:10,25,replace=TRUE)
df<-data.frame(Group,Score1,Score2,Score3,Score4)
library(MASS)
LDA_df=lda(Group~.,data=df)
LDA_df

Output

Call:
lda(Group ~ ., data = df)
Prior probabilities of groups:
  I    II  III   IV
0.24 0.16 0.28 0.32

Group means:
   Score1    Score2   Score3  Score4
I   4.166667 4.333333 5.333333 3.333333
II  4.250000 4.000000 4.000000 2.750000
III 6.000000 5.571429 4.000000 5.000000
IV  5.250000 4.625000 2.125000 5.750000

Coefficients of linear discriminants:
            LD1       LD2       LD3
Score1 0.1358158 0.18645755 -0.18790651
Score2 0.2598956 0.15492088 -0.07433529
Score3 -0.3052648 0.25571648 0.14567716
Score4 0.3117567 0.08656138 0.25216169

Proportion of trace:
  LD1    LD2    LD3
0.8681 0.1161 0.0159

Find the confusion matrix

Create the confusion matrix for linear discriminant analysis performed above by using table and predict function as shown below −

 Live Demo

Group<-sample(c("I","II","III","IV"),25,replace=TRUE)
Score1<-sample(1:10,25,replace=TRUE)
Score2<-sample(1:10,25,replace=TRUE)
Score3<-sample(1:10,25,replace=TRUE)
Score4<-sample(1:10,25,replace=TRUE)
df<-data.frame(Group,Score1,Score2,Score3,Score4)
library(MASS)
LDA_df=lda(Group~.,data=df)
table(predict(LDA_df,type="class")$class,df$Group)

Output

    I   II III IV
I   3    2   1  1
II  2    1   0  0
III 1    0   3  1
IV  0    1   3  6
raja
Nizamuddin Siddiqui
Updated on 13-Aug-2021 15:43:34

Previous Page Print Page Next Page  
Advertisements