Green Lab - Kebab Team - R Package

This repository is part of the project An Empirical Analysis of JavaScript Dead Code in the Wild, described in Kebab Team Base Repository. This repository contains the R package responsible for the experiment design and data analysis.

Getting Started

These instructions will get you a copy of the project up and running on your local machine for experiment replication purpose.

Prerequisites

Tools

• R project
• R Studio IDE (Optional)

R Libraries

• Tidyverse – ggplot2: Provides neat plots, with a more appealing visual and more options to configurate.
• Data Table: Provides more versatile operations with dataframes.

Installation

After installing the tools, you can open and run the scripts in R Studio.

Scripts

The data analysis was split in scripts by domain to improve readability. There are 5 main scripts with additional 6 helper scripts with auxiliary functions.

Main files

• Aggregate all results: Process the raw results from Android Runner located at data/androidrunner/output. It searches the directory for all CSV files in the format Aggregated_Results_Batterystats.csv. Unnecessary columns are dropped and new columns are created with information about the treatment and subject. The formating results are saved at data/androidrunner. All other scripts use the formatted CSV resulted from this script.
• Descriptive Analysis: Provide basic descriptive static of the data.
• Generate Plots: Provide all plots used in the project and the data analysis. ALl generated plots are saved at plots/
• Hypothesis Test: Provide all tests used in this experiment: Shapiro and Kruskal. All results are saved at data\tests\, including a CSV containing a summary of the results.
• Sandbox: Script used for open tests and exploration of the data. To make git stop tracking this file, run git rm --cached sandbox.R

Library scripts

• Constants: Defines all constants used in the scripts.
• IO: Defines all access to read and write data in the file system or the command line.
• Plots: Wraps ggplot2 functions with configurations relevant for this project.
• Preprocessing: Wraps functions for the preprocessing phase, keeping the main file clean and clear.
• Subject: Provides information about subjects.
• Test: Wraps functions for the hypothesis tests.

Authors

• Azim AFroozeh – azimafroozeh;
• Saliha Tabbassum – saalihairshad;
• Stan Swanborn – StanSwanborn;
• Thijmen Kurk – ThijmenKurk;
• Wesley Shann – sshann;

License

This project is licensed under the GLP-3.0 License - see the LICENSE file for details.