Introduction to Data Visualization

Data Visualization Workflow:

Raw Data        →  Cleaned Data  →  Visual Chart  →  Business Insight
(CSV/Database)     (Pandas)         (Matplotlib)      (Decision Made)

Example:
1,200 rows of sales transactions
  ↓ aggregated by month
12 data points (monthly totals)
  ↓ plotted as line chart
"Revenue grew 40% in Q4 — driven by holiday sales"
  ↓ action taken
Marketing increases Q4 budget for next year

Human Processing Speed:
  Reading a table of 1000 numbers:  ~15 minutes
  Viewing a well-designed chart:     ~0.5 seconds

Why our brains prefer visuals:
✅ 65% of humans are visual learners
✅ The human eye can process 36,000 visual messages per hour
✅ Visuals are processed 60,000x faster than text
✅ People remember 80% of what they see vs 20% of what they read

+────────────────────────────────────────────────────────────+
│                    CHART TAXONOMY                          │
+──────────────────+─────────────────────────────────────────+
│ Comparison       │ Bar chart, Grouped bar, Radar           │
│ Trend/Time       │ Line chart, Area chart, Sparkline        │
│ Distribution     │ Histogram, Box plot, Violin, KDE         │
│ Part-of-Whole    │ Pie chart, Donut, Treemap, Waffle        │
│ Relationship     │ Scatter plot, Bubble chart, Heatmap      │
│ Geographic       │ Choropleth map, Dot map, Flow map        │
│ Flow/Process     │ Sankey diagram, Funnel chart             │
│ Statistical      │ Error bars, Confidence intervals         │
+──────────────────+─────────────────────────────────────────+

Rule of Thumb:
  "What do you want to show?"
  Comparison    → Bar chart
  Trend         → Line chart
  Distribution  → Histogram or Box plot
  Part-of-whole → Pie chart (≤5 categories) or Treemap
  Relationship  → Scatter plot
  Geography     → Map

BI Stack:
  Data Source (DB/CSV) → ETL (Extract-Transform-Load) → Data Warehouse
       ↓
  Analytics Layer (Python/SQL)
       ↓
  Visualization Layer (Matplotlib/Seaborn/Plotly/Tableau/Power BI)
       ↓
  Dashboard/Report → Stakeholders → Business Decisions

Key BI Tools:
  Open Source: Matplotlib, Seaborn, Plotly, Grafana, Apache Superset
  Commercial:  Tableau, Power BI, Looker, Qlik

Industry          Use Case                    Chart Used
──────────────────────────────────────────────────────────
Finance           Stock price tracking        Candlestick chart
Healthcare        Patient outcome trends      Line + bar
Retail            Sales by region             Geographic map
Marketing         Funnel conversion rates     Funnel chart
HR                Attrition heatmap           Heatmap
Logistics         Delivery route optimization Flow map
Social Media      Engagement over time        Area chart
Government        Population distribution     Choropleth map
Manufacturing     Quality control             Control chart (SPC)
Sports Analytics  Player performance          Radar/spider chart

import matplotlib.pyplot as plt
import numpy as np

# Student data
students = ['Alice', 'Bob', 'Carol', 'David', 'Eve', 'Frank']
math     = [92, 78, 85, 60, 95, 72]
science  = [88, 82, 90, 55, 97, 65]
english  = [76, 88, 84, 70, 89, 80]

x = np.arange(len(students))
width = 0.25

fig, axes = plt.subplots(1, 2, figsize=(14, 5))

# Chart 1: Grouped bar chart — subject comparison
bars1 = axes[0].bar(x - width, math,    width, label='Math',    color='#2196F3')
bars2 = axes[0].bar(x,          science, width, label='Science', color='#4CAF50')
bars3 = axes[0].bar(x + width,  english, width, label='English', color='#FF9800')
axes[0].set_xticks(x)
axes[0].set_xticklabels(students)
axes[0].set_title('Student Performance by Subject', fontsize=13, fontweight='bold')
axes[0].set_ylabel('Score')
axes[0].set_ylim(0, 110)
axes[0].legend()
axes[0].axhline(y=75, color='red', linestyle='--', alpha=0.6, label='Pass line')

# Chart 2: Average score per student
averages = [(m+s+e)/3 for m,s,e in zip(math, science, english)]
colors = ['#4CAF50' if a >= 80 else '#FF9800' if a >= 70 else '#F44336' for a in averages]
bars = axes[1].barh(students, averages, color=colors)
axes[1].set_title('Overall Average Score', fontsize=13, fontweight='bold')
axes[1].set_xlabel('Average Score')
axes[1].axvline(x=75, color='black', linestyle='--', alpha=0.5)
for bar, avg in zip(bars, averages):
    axes[1].text(avg + 0.5, bar.get_y() + bar.get_height()/2,
                  f'{avg:.1f}', va='center', fontweight='bold')

plt.suptitle('Student Performance Dashboard', fontsize=15, fontweight='bold', y=1.02)
plt.tight_layout()
plt.savefig('student_performance.png', dpi=150, bbox_inches='tight')
plt.show()
print("Chart saved!")