Data Visualization Cheat Sheet — Chart Types & Use Cases for News Data

beginner 8 min read December 29, 2025

data visualizationchart typesmatplotlibplotlynews datapython chartsdata analysisvisualization guidecheat sheetquick reference

🧭

TL;DR: Quick reference for choosing the right chart type when visualizing news data. Covers 4 essential chart types with complete tutorials, working code examples, and NewsDataHub API integration.

This cheat sheet helps you choose the right visualization for your news data analysis. Each chart type includes a complete tutorial with working code, real examples, and NewsDataHub API integration.

---

Quick Selection Guide

Question	Recommended Chart	Tutorial Link
”How many articles per topic/source/country?”	Bar chart	Bar Charts Tutorial
”Which topics co-occur together in articles?”	Chord diagram	Chord Diagram Tutorial
”How do sources distribute across topics?”	Sankey diagram	Sankey Diagram Tutorial
”Show hierarchical topic distribution”	Treemap	Treemap Tutorial

---

Bar Charts

Description: Rectangular bars showing categorical data with bar length representing values.

Best for:

• Topic distributions (articles per topic)
• Source comparisons (most active publishers)
• Language breakdowns
• Country-wise article counts
• Political leaning distribution
• Comparing discrete categories with counts or frequencies

When to use:

• Comparing discrete categories
• Showing rankings or top-N lists
• Displaying counts or frequencies
• Data with 3-20 categories
• When exact values matter more than proportions

When NOT to use:

• Time series data (use line chart instead)
• Showing parts of a whole (use pie chart or treemap)
• Data with 50+ categories (overwhelms visualization)
• Continuous numerical distributions (use histogram)

NewsDataHub use cases:

• “Which topics had the most coverage this week?”
• “Top 10 most active news sources”
• “Article distribution by political leaning”
• “News coverage by country or language”
• “Compare article volume across source types”

Complete Tutorial: How to Create Bar Charts in Python Using Real News Data

Python libraries: matplotlib, seaborn, plotly

Quick example:

First, install matplotlib if you haven’t already:

pip install matplotlib

import requests
import matplotlib.pyplot as plt
from collections import Counter

# Fetch news from NewsDataHub API
response = requests.get(
    'https://api.newsdatahub.com/v1/news',
    headers={
        'X-API-Key': 'YOUR_API_KEY',
        'User-Agent': 'data-visualization-cheat-sheet/1.0-py'
    },
    params={'per_page': 100, 'topic': 'technology'}
)

data = response.json()

# Count articles by topic
topics = [topic for article in data['data'] for topic in article.get('topics', [])]
topic_counts = Counter(topics).most_common(10)

# Create bar chart
plt.figure(figsize=(12, 6))
plt.bar([t[0] for t in topic_counts], [t[1] for t in topic_counts])
plt.xlabel('Topic')
plt.ylabel('Article Count')
plt.title('Top 10 Topics by Article Count')
plt.xticks(rotation=45, ha='right')
plt.tight_layout()
plt.show()

---

Chord Diagrams

Description: Circular diagram showing relationships between entities with connecting ribbons. Ribbon thickness represents connection strength.

Best for:

• Topic co-occurrence (which topics appear together in articles)
• Keyword connections and relationships
• Source-to-topic relationships
• Cross-references between categories
• Many-to-many relationships with flow/connection strength

When to use:

• Showing many-to-many relationships
• Symmetric connections between entities
• Highlighting strongest connections
• Compact visualization of network data
• When you want to emphasize interconnections over individual nodes

When NOT to use:

• One-to-many or hierarchical relationships (use Sankey or treemap)
• When exact numerical values are critical
• Very sparse networks with few connections
• Asymmetric flows (use Sankey diagram instead)

NewsDataHub use cases:

• “Which topics frequently appear together in the same articles?”
• “Keyword co-occurrence patterns in headlines”
• “How topics relate to each other in news coverage”
• “What topics are discussed alongside ‘technology’?”

Complete Tutorial: Chord Diagram for Topic Co-occurrence

Python libraries: holoviews, plotly (custom implementation)

What you’ll learn in the tutorial:

• Fetching articles with multiple topics from NewsDataHub API
• Building co-occurrence matrices from topic data
• Creating interactive chord diagrams with HoloViews
• Interpreting connection patterns in news coverage
• Customizing colors and labels for clarity

---

Sankey Diagrams

Description: Flow diagram showing how data moves between categories. Flow width represents volume.

Best for:

• News flow from sources to topics
• Source type → topic distribution
• Topic → sentiment distribution
• Article journey through categorization stages
• Multi-stage categorical flows with volume

When to use:

• Showing flow or transformation between stages
• Multi-stage categorical data with directionality
• Proportions that move between states
• Illustrating connections AND their relative magnitudes
• When you need to show both direction and volume

When NOT to use:

• Symmetric relationships (use chord diagram)
• Simple hierarchies (use treemap)
• Cyclical flows (Sankey is best for acyclic flows)

NewsDataHub use cases:

• “How do different sources distribute their coverage across topics?”
• “Flow from source political leaning → topics covered”
• “Article sentiment distribution by source type”
• “News coverage flow: source type → topic → sentiment”

Complete Tutorial: Sankey Diagram: News Sources to Topic Flows

Python libraries: plotly

What you’ll learn in the tutorial:

• Aggregating NewsDataHub API data for flow analysis
• Constructing source → topic relationships
• Building Sankey diagrams with Plotly
• Color-coding flows for better readability
• Interactive exploration of news coverage patterns

---

Treemaps

Description: Nested rectangles where size represents hierarchical data. Larger rectangles contain smaller rectangles representing subcategories.

Best for:

• Topic distribution with subtopics
• Hierarchical category breakdown
• Nested categories (e.g., region → country → city)
• Proportional hierarchical data
• Space-efficient visualization of many categories

When to use:

• Hierarchical categorical data
• Showing proportions within categories
• Multiple levels of grouping
• Many categories that need compact visualization
• When you want to see both hierarchy and size at a glance

When NOT to use:

• Flows between categories (use Sankey)
• Non-hierarchical relationships (use chord diagram or network graph)
• When precise numerical comparison is needed (use bar chart)

NewsDataHub use cases:

• “Topic distribution with subtopic breakdown”
• “Source type → specific sources → article counts”
• “Country → region → city news coverage”
• “Hierarchical view of topic taxonomy with article counts”

Complete Tutorial: Treemap Visualization for Topic Distribution

Python libraries: plotly, squarify + matplotlib

What you’ll learn in the tutorial:

• Fetching hierarchical data from NewsDataHub API
• Aggregating article counts by nested categories
• Creating interactive treemaps with Plotly
• Color schemes for hierarchical visualization
• Drill-down interactions for exploration

---

Choosing the Right Chart

Decision Flow

What are you trying to show?

├─ Comparing categories by count/frequency
│  └─ Use: Bar Chart
│     Tutorial: Bar Charts in Python Using Real News Data
│
├─ Which topics appear together?
│  └─ Use: Chord Diagram
│     Tutorial: Chord Diagram for Topic Co-occurrence
│
├─ Flow from one category to another
│  └─ Use: Sankey Diagram
│     Tutorial: Sankey Diagram: News Sources to Topic Flows
│
└─ Hierarchical proportions (e.g., topics → subtopics)
   └─ Use: Treemap
      Tutorial: Treemap Visualization for Topic Distribution

Chart Characteristics Comparison

Chart Type	Shows	Direction	Best Dataset Size	Interactivity
Bar Chart	Categorical counts	None	3-20 categories	Optional
Chord Diagram	Symmetric relationships	Bidirectional	5-20 entities	Recommended
Sankey Diagram	Flows/transformations	Unidirectional	3-30 nodes	Recommended
Treemap	Hierarchical proportions	Hierarchical	Unlimited (nested)	Recommended

---

Best Practices for News Data Visualization

General Guidelines

1. Choose simplicity over complexity

• Start with the simplest chart that answers the question
• Only add complexity when it reveals additional insight
• Bar charts often work better than fancy alternatives

2. Consider your audience

• Journalists/general public: Use familiar charts (bar, pie)
• Data analysts: Use specialized charts (Sankey, chord)
• Provide clear labels and legends for all audiences

3. Label clearly

• Always label axes with units
• Include data source (“Source: NewsDataHub API”)
• Add titles that describe the insight, not just the data
• Use annotations to highlight key findings

4. Use color purposefully

• Color should encode information, not just be decorative
• Use consistent color schemes (same topic = same color across visualizations)
• Ensure accessibility (colorblind-friendly palettes)
• Limit to 7-10 distinct colors maximum

5. Provide context

• Indicate sample size (e.g., “Based on 10,000 articles from Dec 1-15”)
• Show time period clearly
• Compare to benchmarks when meaningful
• Note data limitations (e.g., “48-hour delay for free tier”)

For News Data Specifically

1. Always include time context

• Specify date range clearly
• News is time-sensitive—always show when data was collected
• Compare equivalent time periods when showing trends

2. Normalize when comparing different time periods

• Use percentages or rates rather than absolute counts
• Account for different baseline volumes (e.g., weekend vs weekday)

3. Handle missing data explicitly

• Show “Unknown” or “Other” categories for incomplete data
• Note data quality issues in captions
• Don’t hide missing categories—they tell part of the story

4. Use consistent categorization

• Maintain same topic/source/country groupings across charts
• Use same color schemes throughout a report/dashboard
• Define categories clearly (what counts as “technology” news?)

5. Show source diversity

• When showing topic coverage, indicate number of sources
• Avoid over-representing prolific sources
• Consider weighting by source diversity, not just article count

---

Tools & Libraries Reference

Python Libraries

For Bar Charts:

• matplotlib — Highly customizable, publication-quality
• seaborn — Statistical visualization with beautiful defaults
• plotly — Interactive charts with hover, zoom, export

For Chord Diagrams:

• holoviews — High-level declarative API, integrates with Bokeh
• plotly — Custom implementation possible with advanced features

For Sankey Diagrams:

• plotly — Best Python library for Sankey diagrams
• matplotlib — Limited Sankey support, use plotly instead

For Treemaps:

• plotly — Interactive treemaps with drill-down
• squarify + matplotlib — Static treemaps, good for PDFs/print

Installation

# Install all recommended libraries
pip install matplotlib seaborn plotly holoviews squarify pandas requests

# For Jupyter notebooks
pip install jupyter

Jupyter Notebook Setup

# Standard imports for news data visualization
import requests
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import plotly.graph_objects as go
import plotly.express as px
from collections import Counter

# Configure matplotlib
%matplotlib inline
plt.rcParams['figure.figsize'] = (12, 6)
plt.rcParams['font.size'] = 10

# Configure plotly for notebook
import plotly.io as pio
pio.renderers.default = 'notebook'

---

NewsDataHub API Integration

All visualizations in the tutorials use NewsDataHub API data. Key endpoints:

1. General News Endpoint

import requests

response = requests.get(
    'https://api.newsdatahub.com/v1/news',
    headers={
        'X-API-Key': 'YOUR_API_KEY',
        'User-Agent': 'data-visualization-cheat-sheet/1.0-py'
    },
    params={
        'per_page': 100,
        'topic': 'technology',  # Optional filter
        'language': 'en'        # Optional filter
    }
)

articles = response.json()['data']

2. Extract Visualization Data

For Bar Charts (topic counts):

topics = [topic for article in articles for topic in article.get('topics', [])]
topic_counts = Counter(topics).most_common(10)

For Chord Diagrams (topic co-occurrence):

from itertools import combinations

# Get all topic pairs that co-occur in same articles
pairs = []
for article in articles:
    article_topics = article.get('topics', [])
    if len(article_topics) >= 2:
        pairs.extend(list(combinations(sorted(article_topics), 2)))

co_occurrence = Counter(pairs)

For Sankey Diagrams (source → topic):

flows = []
for article in articles:
    source = article.get('source_title', 'Unknown')
    topics = article.get('topics', ['Unknown'])
    for topic in topics:
        flows.append((source, topic))

flow_counts = Counter(flows)

For Treemaps (hierarchical topics):

# Group by primary topic, then count subtopics
hierarchy = {}
for article in articles:
    topics = article.get('topics', [])
    if topics:
        primary = topics[0]
        if primary not in hierarchy:
            hierarchy[primary] = Counter()
        for subtopic in topics[1:]:
            hierarchy[primary][subtopic] += 1

---

Complete Example Workflow

Here’s a complete workflow from API to visualization:

import requests
import matplotlib.pyplot as plt
from collections import Counter

# Step 1: Fetch data from NewsDataHub API
response = requests.get(
    'https://api.newsdatahub.com/v1/news',
    headers={
        'X-API-Key': 'YOUR_API_KEY',
        'User-Agent': 'data-visualization-cheat-sheet/1.0-py'
    },
    params={
        'language': 'en',
        'per_page': 100
    }
)

articles = response.json()['data']

print(f"Fetched {len(articles)} articles")

# Step 2: Extract topics
topics = []
for article in articles:
    topics.extend(article.get('topics', []))

# Step 3: Count and rank
topic_counts = Counter(topics).most_common(10)

# Step 4: Visualize with bar chart
plt.figure(figsize=(12, 6))
plt.bar(
    [t[0] for t in topic_counts],
    [t[1] for t in topic_counts],
    color='steelblue'
)
plt.xlabel('Topic', fontsize=12)
plt.ylabel('Article Count', fontsize=12)
plt.title('Top 10 Topics in Recent News Coverage', fontsize=14, fontweight='bold')
plt.xticks(rotation=45, ha='right')
plt.tight_layout()

# Add data source annotation
plt.figtext(0.99, 0.01, 'Source: NewsDataHub API',
            ha='right', fontsize=8, style='italic')

plt.show()

# Step 5: Print summary
print("\nTop 10 Topics:")
for topic, count in topic_counts:
    print(f"{topic:20} {count:4} articles")

---

Next Steps

Ready to create your first visualization?

1. Start with bar charts — Easiest to implement, most versatile → Bar Charts Tutorial

2. Explore relationships — Chord diagrams show topic connections → Chord Diagram Tutorial

3. Visualize flows — Sankey diagrams show how news moves between categories → Sankey Diagram Tutorial

4. Show hierarchies — Treemaps compactly display nested categories → Treemap Tutorial

Get your NewsDataHub API key:

• Free tier: 100 requests/day, no credit card required
• Sign up: newsdatahub.com/login
• Docs: newsdatahub.com/docs

---

This cheat sheet focuses on practical, proven chart types with complete working tutorials. Each visualization type has a dedicated guide with real code, NewsDataHub API integration, and example outputs.

Stay Updated

Subscribe to our newsletter to get technical insights, product updates, and behind-the-scenes notes on building NewsDataHub.

Subscribe to Newsletter

Olga S.

Founder of NewsDataHub — Distributed Systems & Data Engineering

Connect on LinkedIn