Exploratory Data Analysis
Machine Learning · Beginner
Definition
EDA summarizes and visualizes data to find patterns, outliers and missing values before modeling.
Bad data understanding causes most ML failures.
Diagram
Flowchart
Example: Mean vs median with outlier
import statistics as s
x = [10, 12, 11, 13, 100]
print(s.mean(x), s.median(x))Output: 29.2 12
Key points
- Median is robust to outliers
- IQR rule flags outliers
Interview questions
IQR outlier rule?
Values below Q1 − 1.5·IQR or above Q3 + 1.5·IQR.