Lesson 1: Introduction to Data in ML & Why Visualization Matters

1. Why Data Matters in Machine Learning

ML models learn from data — garbage in = garbage out.

Real datasets are messy: missing values, outliers, wrong formats, duplicates.

Before modeling → understand, clean, and explore data.

Visualization helps spot patterns, errors, and relationships humans can’t see in numbers alone.

2. Key Goals of Data Handling

• Load data from files (CSV, Excel, JSON)
• Clean: fix missing values, remove duplicates
• Explore: stats, distributions, correlations
• Transform: normalize, encode categories
• Visualize: plots to understand features & target

Tools: Pandas (data manipulation), Matplotlib/Seaborn (visualization)

Summary – Lesson 1

• Data preparation is the biggest part of ML
• Visualization reveals patterns, errors, relationships
• Next: Pandas Series (1D data building block)