Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update 05-support-vector-machines.Rmd #3

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Update 05-support-vector-machines.Rmd #3

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
47 changes: 46 additions & 1 deletion 05-support-vector-machines.Rmd
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -255,7 +255,7 @@ plot(svmTune, metric = "ROC", scales = list(x = list(log =2)))
Predictions on test set.
```{r echo=T}
svmPred <- predict(svmTune, moonsTest[,c(1:2)])
confusionMatrix(svmPred, as.factor(moonsTest[,3]))
confusionMatrix(svmPred, moonsTest[,3])
```

Get predicted class probabilities so we can build ROC curve.
Expand Down Expand Up @@ -336,6 +336,51 @@ ggplot(xgrid, aes(V1,V2)) +
axis.title=element_text(size=20,face="bold"))
```

**Iris example**

splitting the data into training set and test set
```{r echo=T}
library(e1071)
library(caTools)
my.split = sample.split(iris$Species, SplitRatio = .8)
training_set = subset(iris, my.split == TRUE)
test_set = subset(iris, my.split == FALSE)
nrow(training_set)
```

```{r echo=T}
training_set[,1:4] = scale(training_set[,1:4])
test_set[,1:4] = scale(test_set[,1:4])
classifier1 = svm(formula = Species~., data = training_set, type = 'C-classification', kernel = 'radial')
classifier2 = svm(formula = Species~ Petal.Width + Petal.Length, data = training_set, type = 'C-classification', kernel = 'radial')
```


```{r echo=T}
test_pred1 = predict(classifier1, type = 'response', newdata = test_set[-5])
test_pred2 = predict(classifier2, type = 'response', newdata = test_set[-5])
# Making Confusion Matrix
cm1 = table(test_set[,5], test_pred1)
cm2 = table(test_set[,5], test_pred2)
cm1 # Confusion Matrix for all parameters
cm2 # Confusion Matrix for parameters being Petal Length and Petal Width
```

The accuracy for both model looks solid. Also notice that as we had deduced, only Petal Length and Width is important to make this model accurate and our second classifier proves it!

```{r svm.plots}
m2 <- svm(Species~., data = iris)
plot(m2, iris, Petal.Width ~ Petal.Length,
slice = list(Sepal.Width = 3, Sepal.Length = 4))
```

```{r svm.subsets}
iris.part = subset(iris, Species != 'setosa')
iris.part$Species = factor(iris.part$Species)
#iris.part = iris.part[, c(1,2,5)]
svm.fit = svm(formula=Species~., data=iris.part, type='C-classification', kernel='linear')
plot(svm.fit, iris.part, Petal.Width ~ Petal.Length, slice = list(Sepal.Width = 3, Sepal.Length = 4))
```

## Example - regression
This example serves to demonstrate the use of SVMs in regression, but perhaps more importantly, it highlights the power and flexibility of the [caret](http://cran.r-project.org/web/packages/caret/index.html) package. Earlier we used _k_-NN for a regression analysis of the **BloodBrain** dataset (see section \@ref(knn-regression)). We will repeat the regression analysis, but this time we will fit a radial kernel SVM. Remarkably, a re-run of this analysis using a completely different type of model, requires changes to only two lines of code.
Expand Down