BioSeqClass (version 1.30.0)

selectFFS: feature forward selection

Description

feature forward selection.

Usage

selectFFS(data, accCutoff, stop.n, classifyMethod="knn",cv=10)

Arguments

data

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.

accCutoff

a numeric indicating the minimum difference of accuracy between two models in selectFFS. Feature subsets will stop increasing when the difference of accuracy is samll than accCutoff.

stop.n

number of selected features by selectFFS.

classifyMethod

a string for the classification method. This must be one of the strings "libsvm", "svmlight", "NaiveBayes", "randomForest", "knn", "tree", "nnet", "rpart", "ctree", "ctreelibsvm", "bagging".

cv

an integer for the time of cross validation, or a string "leave\_one\_out" for the jacknife test.

Details

selectFFS uses FFS (Feature Forword Selection) method to increase feature, and use NNA (Neareast Neighbor Analysis) to evaluate the performance of feature subset. Two conditions are used to stop feature increasing: control the difference of accuracy between two models; control the number of selected features by Parameter "stop.n".

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]
  seq=c(posSeq,negSeq)
  classLable=c(rep("+1",length(posSeq)),rep("-1",length(negSeq)) ) 
  data = data.frame(featureBinary(seq),classLable)
  
  if(interactive()){  
    ## Use KNN to evaluate the performance of feature subset, 
    ## and use Feature Forword Selection method to increase feature.
    # If the difference of accuracy between two models is less than 0.01, feature 
    # selection will stop.
    FFS_NNA_CV5 = selectFFS(data,accCutoff=0.01,classifyMethod="knn",cv=5)
    # If 20 features have been selected, feature selection will stop.
    FFS_NNA_CV5 = selectFFS(data,stop.n=3,classifyMethod="knn",cv=5)
    # If any one condiction is satisfied, feature selection will stop.
    FFS_NNA_CV5 = selectFFS(data,accCutoff=0.001,stop.n=100,classifyMethod="knn",cv=5)   
  }