BioSeqClass (version 1.30.0)

model: Classification Models

Description

These functions build various classification models.

Usage

classifyModelLIBSVM(train,svm.kernel="linear",svm.scale=FALSE) classifyModelSVMLIGHT(train,svm.path,svm.options="-t 0") classifyModelNB(train) classifyModelRF(train) classifyModelKNN(train, test, knn.k=1) classifyModelTree(train) classifyModelNNET(train, nnet.size=2, nnet.rang=0.7, nnet.decay=0, nnet.maxit=100) classifyModelRPART(train) classifyModelCTREE(train) classifyModelCTREELIBSVM(train, test, svm.kernel="linear",svm.scale=FALSE) classifyModelBAG(train)

Arguments

train

a data frame including the feature matrix and class label. The last column is a vector of class label comprising of "-1" or "+1"; Other columns are features.

svm.kernel

a string for kernel function of SVM.

svm.scale

a logical vector indicating the variables to be scaled.

svm.path

a character for path to SVMlight binaries (required, if path is unknown by the OS).

svm.options

Optional parameters to SVMlight. For further details see: "How to use" on http://svmlight.joachims.org/. (e.g.: "-t 2 -g 0.1"))

nnet.size

number of units in the hidden layer. Can be zero if there are skip-layer units.

nnet.rang

Initial random weights on [-rang, rang]. Value about 0.5 unless the inputs are large, in which case it should be chosen so that rang * max(|x|) is about 1.

nnet.decay

parameter for weight decay.

nnet.maxit

maximum number of iterations.

knn.k

number of neighbours considered in function classifyModelKNN.

test

a data frame including the feature matrix and class label. The last column is a vector of class label comprising of "-1" or "+1"; Other columns are features.

Details

classifyModelLIBSVM builds support vector machine model by LibSVM. R package "e1071" is needed. classifyModelSVMLIGHT builds support vector machine model by SVMlight. R package "klaR" is needed. classifyModelNB builds naive bayes model. R package "klaR" is needed.

classifyModelRF builds random forest model. R package "randomForest" is needed. classifyModelKNN builds k-nearest neighbor model. R package "class" is needed. classifyModelTree builds tree model. R package "class" is needed.

classifyModelRPART builds recursive partitioning trees model. R package "rpart" is needed. classifyModelCTREE builds conditional inference trees model. R package "party" is needed. classifyModelCTREELIBSVM combines conditional inference trees and support vecotr machine. R package "party" and "e1071" is needed. classifyModelBAG uses bagging method. R package "ipred" is needed.

Examples

  ## read positive/negative sequence from files.
  tmpfile1 = file.path(path.package("BioSeqClass"), "example", "acetylation_K.pos40.pep")
  tmpfile2 = file.path(path.package("BioSeqClass"), "example", "acetylation_K.neg40.pep")
  posSeq = as.matrix(read.csv(tmpfile1,header=FALSE,sep="\t",row.names=1))[,1]
  negSeq = as.matrix(read.csv(tmpfile2,header=FALSE,sep="\t",row.names=1))[,1]
  data = data.frame(rbind(featureBinary(posSeq,elements("aminoacid")), 
       featureBinary(negSeq,elements("aminoacid")) ),
       class=c(rep("+1",length(posSeq)),
       rep("-1",length(negSeq))) )
  
  ## sample train and test data
  tmp = c(sample(1:length(posSeq),length(posSeq)*0.8), 
    sample(length(posSeq)+(1:length(negSeq)),length(negSeq)*0.8))
  train = data[tmp,]
  test = data[-tmp,]
  
  ## Build classification model using training data
  model1 = classifyModelLIBSVM(train,svm.kernel="linear",svm.scale=FALSE)
  ## Predict test data by classification model
  testClass = predict(model1, test[,-ncol(test)])