Data%20Mining%20Packages%20in%20R:%20logistic%20regression%20and%20SVM

1
Data Mining Packages in R logistic regression
and SVM

• Jiang Du
• March 2008

2
Logistic Regression

• lrm in package Design
• http//biostat.mc.vanderbilt.edu/s/Design/html/lrm
  .html
• glm in package stats
• http//finzi.psych.upenn.edu/R/library/stats/html/
  glm.html

3
Logistic Regression lrm

• Usage
• lrm(formula, data, subset, na.actionna.delete,
  method"lrm.fit", modelFALSE, xFALSE, yFALSE,
  linear.predictorsTRUE, se.fitFALSE, penalty0,
  penalty.matrix, tol1e-7, strata.penalty0,
  var.penaltyc('simple','sandwich'), weights,
  normwt, ...)
• Arguments
• Formula
• a formula object. An offset term can be included.
  The offset causes fitting of a model such as
  logit(Y1) Xß W, where W is the offset
  variable having no estimated coefficient. The
  response variable can be any data type lrm
  converts it in alphabetic or numeric order to an
  S factor variable and recodes it 0,1,2,...
  internally.
• Data
• data frame to use. Default is the current frame.

• Usage
• S3 method for class 'lrm'
• predict(object, ..., typec("lp", "fitted",
  "fitted.ind", "mean", "x", "data.frame", "terms",
  "adjto","adjto.data.frame", "model.frame"),
  se.fitFALSE, codesFALSE)
• Arguments
• Object
• a object created by lrm
• ...
• arguments passed to predict.Design, such as kint
  and newdata (which is used if you are predicting
  out of data). See predict.Design to see how NAs
  are handled.
• Type

4
Logistic Regression lrm

• Fitting training data
• model lrm(Class X Y Z, datatrain)
• Prediction on new data
• To get logit(Y1)
• predict(model, newdata test, type lp)
• To get Pr(Y1)
• predict(model, newdata test, type
  fitted.ind)

5
?formula

• The models fit by, e.g., the lm and glm functions
  are specified in a compact symbolic form. The
  operator is basic in the formation of such
  models. An expression of the form y model is
  interpreted as a specification that the response
  y is modelled by a linear predictor specified
  symbolically by model. Such a model consists of a
  series of terms separated by operators. The
  terms themselves consist of variable and factor
  names separated by operators. Such a term is
  interpreted as the interaction of all the
  variables and factors appearing in the term.
• In addition to and , a number of other
  operators are useful in model formulae. The
  operator denotes factor crossing ab interpreted
  as abab. The operator indicates crossing to
  the specified degree. For example (abc)2 is
  identical to (abc)(abc) which in turn
  expands to a formula containing the main effects
  for a, b and c together with their second-order
  interactions. The in operator indicates that
  the terms on its left are nested within those on
  the right. For example a b in a expands to
  the formula a ab. The - operator removes the
  specified terms, so that (abc)2 - ab is
  identical to a b c bc ac. It can also
  used to remove the intercept term y x - 1 is a
  line through the origin. A model with no
  intercept can be also specified as y x 0 or y
  0 x.

6
Logistic Regression glm

• Fitting training data
• model glm(Class X Y Z, datatrain,
  familybinomial(logit))
• Prediction on new data
• To get logit(Y1)
• predict(model, newdata test)
• To get Pr(Y1)
• predict(model, newdata test, type response)

7
SVM

• svm in e1071
• http//www.potschi.de/svmtut/svmtut.html
• ksvm in kernlab
• http//rss.acs.unt.edu/Rdoc/library/kernlab/html/k
  svm.html

8
SVM svm

• Kernel
• the kernel used in training and predicting. You
  might consider changing some of the following
  parameters, depending on the kernel type.
• linear
• u'v
• polynomial
• (gammau'v coef0)degree
• radial basis
• exp(-gammau-v2)
• sigmoid
• Tanh(gammau'v coef0)

9
SVM svm

• Training
• model svm(Class X Y Z, datatrain, type
  "C, kernel linear)
• Prediction
• predict(model, newdata test)