scripts/builtin/outlierByIQR.dml

#-------------------------------------------------------------
#
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements.  See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership.  The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License.  You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied.  See the License for the
# specific language governing permissions and limitations
# under the License.
#
#-------------------------------------------------------------


# Builtin function for detecting and repairing outliers using standard deviation 
#
# INPUT PARAMETERS:
# ---------------------------------------------------------------------------------------------
# NAME            TYPE    DEFAULT     MEANING
# ---------------------------------------------------------------------------------------------
# X               Double   ---       Matrix X  
# k               Double   1.5       a constant used to discern outliers k*IQR 
# isIterative     Boolean  TRUE      iterative repair or single repair 
# repairMethod    Integer  1         values: 0 = delete rows having outliers, 
#                                            1 = replace outliers with zeros 
#                                            2 = replace outliers as missing values 
# max_iterations  Integer  0         values: 0 = arbitrary number of iteraition until all outliers are removed, 
#                                    n = any constant defined by user
# ---------------------------------------------------------------------------------------------


#Output(s)
# ---------------------------------------------------------------------------------------------
# NAME            TYPE    DEFAULT     MEANING
# ---------------------------------------------------------------------------------------------
# Y               Double   ---       Matrix X with no outliers

m_outlierByIQR = function(Matrix[Double] X, Double k =1.5, Integer repairMethod = 1, 
  Integer max_iterations, Boolean verbose = TRUE) return(Matrix[Double] Y) 
{
  sumPrevious =0
  sumNext = 1
  counter = 0

  while( max_iterations == 0 | counter < max_iterations )
  {
    [Q1, Q3, IQR] = compute_quartiles(X)
    upperBound = (Q3 + (k * IQR));
    lowerBound = (Q1 - (k * IQR));
    outlierFilter = X < lowerBound | X > upperBound
    if(sum(outlierFilter) > 1 & sumNext != 0 & sumPrevious != sumNext ) {
      #TODO: see outlierBySd why are sumPrevious and sumNext necessary    	
      temp = replace(target=X, pattern = NaN, replacement = 0)
      sumPrevious = sum(temp)
      X = fix_outliers_iqr(X, outlierFilter, repairMethod)
      temp = replace(target=X, pattern = NaN, replacement = 0)
      sumNext = sum(temp)
    }
    else
      max_iterations = -1

    counter = counter + 1; 
  }
  Y = X
  if(verbose) {
    print("Total executed iterations = "+counter)
    print("Upper-bound of data was calculated using Q3 + k * IQR")
    print("lower-bound of data was calculated using Q3 - k * IQR")
    print("Anything less than the lower-bound and greater than the upper-bound was treated as outlier")
    if(sum(Y) == 0)
      print("output is a zero matrix due to iterative evaluation of outliers ")
    print("output:\n"+ toString(Y))
  }
}
  
fix_outliers_iqr = function(Matrix[Double] X, Matrix[Double] outlierFilter, Integer repairMethod = 1)
  return(Matrix[Double] fixed_X)
{
  rows = nrow(X)
  cols = ncol(X)
  if(repairMethod == 0) {
    sel = rowMaxs(outlierFilter) == 0
    X = removeEmpty(target = X, margin = "rows", select = sel)
  }
  else if(repairMethod == 1)
    X = (outlierFilter == 0) * X
  else if(repairMethod == 2)
  {
    outlierFilter = replace(target = (outlierFilter == 0), pattern = 0, replacement = NaN)
    X = outlierFilter * X
  }
  else
    stop("outlierByIQR: invalid argument - repair required 0-2 found: "+repairMethod)

  fixed_X = X
}

compute_quartiles = function(Matrix[Double] X)
  return(Matrix[Double] colQ1, Matrix[Double] colQ3, Matrix[Double] IQR)
{
  cols = ncol(X)
  colQ1 = matrix(0, 1, cols)
  colQ3 = matrix(0, 1, cols)
  if(nrow(X) > 1) {
    parfor(i in 1:cols) {
      colQ1[,i] = quantile(X[,i], 0.25)
      colQ3[,i] = quantile(X[,i], 0.75)
    }
  }
  IQR = colQ3 - colQ1
}