R-Software SVM prediction example

Student	x	y	Grade
1	4	85	B
2	6	80	B
3	8	92	A
4	4	70	C
5	2	65	C
6	1	60	C
7	5	89	B
8	7	82	B
9	4	81	B
10	6	95	A

-----------------------
Add library

library(e1071)

-----------------------

Read the data from csv file

a<-read.csv(file=file.choose(), header=TRUE)

-----------------------

name the columns

names(a)

------------------------

partition the data into testset and trainset

> index <-1:nrow(a)

> testindex <-sample(index,trunc(length(index)*30/100))

> testset <-a[testindex,]

> trainset<-a[-testindex,]

-----------------------------

label them

> x2<-trainset[,2]

> y2<-trainset[,3]

> z2<-trainset[,4]

> x1<-testset[,2]

> y1<-testset[,3]

> z1<-testset[,4]

-------------------------------

Create the dataframe

DF <-data.frame(x=x2,y=y2,z=z2)

---------------------------------

Train the model

 mod <-svm(z~x,data=DF,kernel="linear")

------------------------------

Predict

prediction <-predict(mod, newdata=data.frame(x=x1))

----------------------

Table out against the true values

 tab <-table(pred=prediction,true=z1)

----------------------------------

Result

  true

pred A B C

   A 0 0 0

   B 1 1 0

   C 0 0 1

--------------------------------
Library(kernlab)
Least Square SVM
 mod <-lssvm(z~x,data=DF,kernel="rbfdot")

--------------------------------
Least Square SVM kernal trick


> rbf=rbfdot(15)                                               ; 15 is the value of sigma, kernal scaling parameter
>  mod <-lssvm(z~x,data=DF,kernel=rbf )

----------------------------------------------

predict example

library(e1071)

x <- c(1:10)
y <- c(0,0,0,0,1,0,1,1,1,1)
test <- c(11:15)

> DF <- data.frame(x = x, y = y)
> mod <- svm(y ~ x, data = DF, kernel = "linear", gamma = 1, cost = 2,
+ type="C-classification")
> predict(mod, newdata = data.frame(x = test))
1 2 3 4 5 
1 1 1 1 1 
Levels: 0 1

Reading from CSV file

 a <-read.csv(file.choose(), header=TRUE)

One can give path of file instead of file.choose()

Extrating sql data to CSV file

[11/9/2012 3:35:43 PM] Osman: SELECT book_id , name , author  
INTO OUTFILE 'c:/abc.csv'
FIELDS TERMINATED BY ','
ENCLOSED BY '\"'
LINES TERMINATED BY '\n'
FROM books