Skip to content

In svm(), how to connect quantitative prediction result to categorical result?

4 messages · Li, Yunfei, Saeed Abu Nimeh, Steve Lianoglou

2 days later
Steve Lianoglou
# 
Hi Yunfei,
On Fri, Apr 8, 2011 at 8:35 PM, Li, Yunfei <yunfei_li at wsu.edu> wrote:
You have to figure out if you want the svm to do classification or regression.

If I remember correctly, a "vanilla" call to SVM will try to pick one
or the other in a "smart way" by looking at the types (and values) of
your labels (y vector).

You can be more explicit and tell the SVM what you want by specifying
a value for the `type` argument in your original `svm` call.

See ?svm for more info.

I'm not sure if I'm answering your question or not(?). If I didn't
understand what you wanted, perhaps you can rephrase your question, or
maybe explain how my answer is not what you were after ... otherwise
hopefully someone else can provide a better answer.

-steve
1 day later
Saeed Abu Nimeh
# 
I trained a linear svm and did classification. looking at the model I
have, with a binary response 0/1, the decision values look like this:
head(svm.model$decision.values)
2.5
3.1
-1.0

looking at the fitted values
head(svm.model$fitted)
1
1
0
So it looks like anything less than or equal 0 is mapped to the
negative class, i.e. 0), otherwise it is mapped to the positive class,
i.e. 1.
On Fri, Apr 8, 2011 at 8:35 PM, Li, Yunfei <yunfei_li at wsu.edu> wrote:
Steve Lianoglou
# 
Hi,
On Tue, Apr 12, 2011 at 10:54 AM, Saeed Abu Nimeh <sabunime at gmail.com> wrote:
Yes -- so far, so good.

In SVM classification, when examples are predicted with a positive
decision value they are assigned to one class (lets say +1), and
examples with negative decision value are assigned to the other (-1).

Was there a remaining question, or?

-steve