OEdii

Overview

OEdii: Outlier Elimination Technique Using Deletion - Imputation Iteration

OEdii is R package and outlier elimination technique using deletion-imputation iteration.

It combines prodNA (by missForest package) and Multiple Imputation Techniques (Mice and missForest in this version) to delete outlier cases.

See help text (in the making) to know more information and procedure of outlier deletion. In this version, dataset must be constructed by ONLY numeric veriables.

Installation

devtools::install_github("koji-to/OEdii")

Dependencies

mice

missForest

Usage

OEdii(
  data.df,
  imp = "mice",
  del_rate = 0.1,
  elim_rate = 0.2,
  iter = 1000,
  penl = "SQD"
)

Arguments

data.df target dataset

imp imputation method name ('mice' (default setting) or 'missForest')

del_rate deletion rate to complete dataset

elim_rate outlier elimination rate

iter the number of of iteration

penl calculation method of residual ('SQD': Squared difference (default setting) or 'ABD': Absolute difference)

Values

$elim returns dataframe without outlier (eliminated)

$diff returns means of difference between imputed value and actual one about each field

$mcont returns the number of missing count of each field

Example

library(OEdii)

OEdic(iris[,-ncol(iris)], iter=100)
out <- OEdic(iris[,-ncol(iris)],iter=100)
out$elim

Citation

@article{https://doi.org/10.1002/tee.23889,

author = {Toda, Koji and Tsunoda, Masateru},

title = {Outlier Elimination Technique Using Deletion-Imputation Iteration for Fault-Prone Module Detection},

journal = {IEEJ Transactions on Electrical and Electronic Engineering},

volume = {18},

number = {10},

pages = {1653-1663},

keywords = {outlier elimination, missing value imputation, fault-prone detection},

doi = {https://doi.org/10.1002/tee.23889},

url = {https://onlinelibrary.wiley.com/doi/abs/10.1002/tee.23889},

eprint = {https://onlinelibrary.wiley.com/doi/pdf/10.1002/tee.23889},

abstract = {In software development, improving the efficiency of a testing process is important to ensure reliability during a limited development period. One approach to improving the testing process is identifying modules that are likely to contain faults (fault-prone modules) and allocate effort toward resolving them. For this purpose, many fault-prone module detection models have been proposed in previous studies. However, an appropriate fault-prone module detection model cannot be constructed if outliers, such as modules with a significantly large number of source code lines and branches, but no faults, are included. In this study, we propose a new outlier elimination technique that creates missing values (deleted values) to complete the data artificially and applies a missing value imputation technique using a regression approach. If the imputed value differs significantly from the actual (recorded) value, the proposed technique treats the values as outliers. We name the proposed technique Outlier Elimination Technique using Deletion-Imputation Iteration (OEdii), and its performance is verified experimentally. In general, our experimental results are more accurate than the previous outlier elimination techniques in the area under the ROC curve. © 2023 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.}

}