DataTerminology.comNEW
Also known as: EDA
A preliminary analysis approach that uses descriptive statistics and visualizations to understand data structure, spot anomalies, and generate hypotheses.
DESCRIPTION
Exploratory Data Analysis emphasizes graphing and summarizing data before modeling. Originating with John Tukey, EDA reveals patterns, outliers, and relationships that inform feature engineering and modeling choices. Techniques include histograms, box plots, scatterplots, correlation matrices, and profiling. EDA is iterative and question-driven, often performed in notebooks. It reduces model risk by validating assumptions and identifying data issues early. While complementary to confirmatory analysis, EDA avoids rigid hypotheses in favor of discovery.
ALTERNATIVE DEFINITIONS
Wikipedia
Exploratory data analysis (EDA) is an approach to analyzing data sets to summarize their main characteristics, often with visual methods.
Stresses summarization and visual methods as core to EDA.
IBM
Exploratory data analysis is the process of examining data sets to summarize their main characteristics, often using statistical graphics and other data visualization methods.
Connects EDA to practical visualization techniques for insights.
Microsoft Learn
Exploratory data analysis involves visualizing and transforming data to identify patterns, detect anomalies, and test hypotheses before modeling.
Places EDA explicitly in the ML workflow for hypothesis testing.
RELATED TERMS
Visualization Descriptive Statistics Data Anomaly Detection Hypothesis Generation Data Profiling Feature Engineering Statistical Graphics Feature Reduction Sampling Scrubbing
COMPETING TERMS
KNOWLEDGE AREAS
Data Science Data Visualization Testing & Validation Data Quality
TARGET AUDIENCE
Data Scientist Data Analyst BI Specialist AI Engineer Information Analyst
EXTERNAL SOURCES
- Wikipedia - Exploratory data analysis
- IBM - EDA overview and techniques
- Microsoft - EDA in ML lifecycle guidance
- Wikipedia - Historical background and concepts
TOP QUESTIONS
Why is Exploratory Data Analysis crucial before modeling?
EDA uncovers data issues and patterns that shape feature choices and model selection, reducing downstream errors.
Category: Relevancy
What tools are popular for Exploratory Data Analysis?
Common choices include Python (pandas, seaborn), R (ggplot2), and BI tools for quick visual summaries.
Category: Tools
What are common mistakes in Exploratory Data Analysis?
Overfitting insights to small samples, ignoring data leakage, and failing to validate patterns with holdouts are frequent pitfalls.
Category: Mistakes
How does Exploratory Data Analysis differ from Confirmatory Data Analysis?
EDA is discovery-oriented and iterative; CDA tests predefined hypotheses with formal statistical procedures.
Category: Comparison
What is a hot topic in Exploratory Data Analysis?
Automated EDA and profiling tools accelerate discovery while ensuring reproducibility with code-first workflows.
Category: Topic
How much time should be spent on Exploratory Data Analysis?
Practitioners often spend substantial time—sometimes half—on EDA and data preparation to ensure robust modeling.
Category: Practice
Can Exploratory Data Analysis be fully automated?
Automation helps, but human judgment is essential for contextual interpretation and avoiding misleading artifacts.
Category: Opinion
How does Exploratory Data Analysis support data quality?
It reveals missingness, outliers, and inconsistencies, informing cleaning and validation steps.
Category: Relationship
Where should newcomers start with Exploratory Data Analysis?
Begin with univariate plots, basic statistics, and pairwise charts, then iterate with domain questions.
Category: GettingStarted