README

AutoEDA is an R package for performing comprehensive Automatic Exploratory Data Analysis (EDA) with minimal code.

The package generates descriptive statistics, missing value summaries, visualizations, correlation analysis, outlier detection, principal component analysis (PCA), clustering, and automated reports.

Features

Installation

# install.packages("remotes")

remotes::install_github("vinodhpmd/AutoEDA")

Quick Start

library(AutoEDA)

report <- auto_eda(iris)

report
#> 
#> ========================================
#>         AutoEDA Report
#> ========================================
#> 
#> Modules Completed
#> 
#> * Summary 
#> * Missing 
#> * Numeric 
#> * Categorical 
#> * Correlation 
#> * Outliers 
#> * PCA 
#> * Cluster

Dataset Summary

summary_data(iris)
#> $Rows
#> [1] 150
#> 
#> $Columns
#> [1] 5
#> 
#> $NumericVariables
#> [1] 4
#> 
#> $CharacterVariables
#> [1] 0
#> 
#> $FactorVariables
#> [1] 1
#> 
#> $LogicalVariables
#> [1] 0
#> 
#> $MissingValues
#> [1] 0
#> 
#> $DuplicateRows
#> [1] 1
#> 
#> $MemoryMB
#> [1] 0.01
#> 
#> attr(,"class")
#> [1] "SummaryData"

Missing Value Analysis

missing_summary(iris)
#> 
#> =========================================
#>       AutoEDA Missing Value Report
#> =========================================
#> 
#> Rows                         150
#> Columns                      5
#> Variables with Missing       0
#> Complete Cases               150
#> Total Missing Values         0
#> Overall Missing              0.00%
#> 
#> Variable Summary
#> -----------------------------------------
#>      Variable    Type Missing Percent Complete
#>  Sepal.Length numeric       0       0      150
#>   Sepal.Width numeric       0       0      150
#>  Petal.Length numeric       0       0      150
#>   Petal.Width numeric       0       0      150
#>       Species  factor       0       0      150

Numeric Summary

numeric_summary(iris)
#> 
#> =========================================
#>      AutoEDA Numeric Summary
#> =========================================
#> 
#>      Variable   N Missing Mean Median    SD Variance     SE   CV Minimum  Q1
#>  Sepal.Length 150       0 5.84   5.80 0.828    0.686 0.0676 14.2     4.3 5.1
#>   Sepal.Width 150       0 3.06   3.00 0.436    0.190 0.0356 14.3     2.0 2.8
#>  Petal.Length 150       0 3.76   4.35 1.765    3.116 0.1441 47.0     1.0 1.6
#>   Petal.Width 150       0 1.20   1.30 0.762    0.581 0.0622 63.6     0.1 0.3
#>   Q3 Maximum IQR Range Skewness Kurtosis Shapiro_P
#>  6.4     7.9 1.3   3.6    0.309   -0.606  1.02e-02
#>  3.3     4.4 0.5   2.4    0.313    0.139  1.01e-01
#>  5.1     6.9 3.5   5.9   -0.269   -1.417  7.41e-10
#>  1.8     2.5 1.5   2.4   -0.101   -1.358  1.68e-08

Correlation Analysis

correlation_analysis(iris)
#> 
#> =====================================
#>       Correlation Matrix
#> =====================================
#> 
#>              Sepal.Length Sepal.Width Petal.Length Petal.Width
#> Sepal.Length        1.000      -0.118        0.872       0.818
#> Sepal.Width        -0.118       1.000       -0.428      -0.366
#> Petal.Length        0.872      -0.428        1.000       0.963
#> Petal.Width         0.818      -0.366        0.963       1.000

Principal Component Analysis

pca <- pca_analysis(iris)

pca
#> 
#> ==============================
#>  Principal Component Analysis
#> ==============================
#> 
#>          Length Class  Mode   
#> sdev       4    -none- numeric
#> rotation  16    -none- numeric
#> center     4    -none- numeric
#> scale      4    -none- numeric
#> x        600    -none- numeric

Cluster Analysis

cluster_analysis(iris)
#> 
#> ===================================
#>       Cluster Analysis
#> ===================================
#> 
#> Clusters : 3 
#> 
#> Cluster Sizes
#> [1] 53 47 50

Generate Complete Report

report <- auto_eda(iris)

report
#> 
#> ========================================
#>         AutoEDA Report
#> ========================================
#> 
#> Modules Completed
#> 
#> * Summary 
#> * Missing 
#> * Numeric 
#> * Categorical 
#> * Correlation 
#> * Outliers 
#> * PCA 
#> * Cluster

AutoEDA

Overview