curveCtuPlot()

Plot Temporal Curves with Highlighted Example Keywords by Zone

The curveCtuPlot() function creates a temporal visualization that combines zone-based frequency curves with highlighted example keywords. This allows you to show both the overall zone structure and specific representative terms simultaneously, making it ideal for presentations and publications.

---

🔹 Function Definition

curveCtuPlot(
  data,
  ctu_noun = NULL,
  r = 1,
  themety = "light",
  size_class = NULL,
  x_lab = "years"
)

---

🎯 Purpose

While curvePlot() shows all keywords together, curveCtuPlot() provides a more structured visualization by combining zone-level patterns with specific example keywords. This function helps you:

1. Show zone structure — Display aggregate patterns for each frequency zone
2. Highlight representatives — Emphasize specific keywords as examples
3. Create clear narratives — Tell a story through selected exemplar terms
4. Combine overview and detail — Balance big picture with specific cases
5. Improve readability — Reduce clutter while maintaining informativeness
6. Support presentations — Create clear, focused visualizations for talks
7. Guide interpretation — Direct attention to key terms within zones
8. Demonstrate methodology — Show how zone classification works with concrete examples

This function is particularly useful when you want to demonstrate temporal patterns without overwhelming your audience with hundreds of trajectories.

---

🧮 Understanding CTU (Characteristic Terms by Usage)

CTU refers to selecting representative keywords from different frequency zones: - High-frequency zone: Core vocabulary terms - Medium-frequency zone: Specialized but common terms
- Low-frequency zone: Rare or emerging terms

By highlighting one keyword from each zone, you show the spectrum of vocabulary in your corpus.

---

⚙️ Arguments

Argument	Type	Default	Description
data	List	required	A list object returned by importData() or normalization(), containing the TDM and corpus metadata.
ctu_noun	Character vector	NULL	Vector of exactly 3 keywords to highlight, one from each frequency zone (typically high, medium, low). Order doesn’t matter; function assigns them to zones automatically. If NULL, function may auto-select representative terms.
r	Integer	1	Interval for x-axis label thinning. r = 1 shows all years, r = 2 shows every 2nd year, etc.
themety	Character	"light"	Visual theme: • "light": Light background (default) • "dark": Dark background (presentations)
size_class	Numeric vector	NULL	Custom line thickness for the three main zones (typically high, medium, low). If NULL, uses theme-appropriate defaults. Vector length should match number of main zones.
x_lab	Character	"years"	Label for x-axis (e.g., "Years", "Time Period", "Publication Year").

---

📊 Plot Components

Visual Structure

The plot contains two layers:

Layer 1: Zone Curves (Background)

• Lines representing aggregated trajectories for each frequency zone
• Colors indicate zones (using corpus color palette)
• Line thickness varies by zone importance
• Shows overall zone behavior

Layer 2: Example Keywords (Foreground)

• 3 highlighted keywords from different zones
• Distinct colors separate from zone colors
• Emphasized line thickness for visibility
• Labels in legend identify specific terms

Dual Legend System

Zone Legend: - Shows frequency zones with interval ranges - Example: “Zone 4 [500-1000]”, “Zone 2 [50-100]”

Keyword Legend: - Shows the 3 highlighted example keywords - Example: “algorithm”, “preprocessing”, “noise”

---

💡 Usage Examples

Basic Usage with Selected Keywords

library(cccc)

# Import and normalize data
corpus <- importData("tdm.csv", "corpus_info.csv")
corpus_norm <- normalization(corpus, normty = "nc")

# Plot with 3 representative keywords from different zones
curveCtuPlot(
  corpus_norm,
  ctu_noun = c("algorithm", "preprocessing", "noise")
)

Automatic Keyword Selection

# Let function auto-select representative keywords
curveCtuPlot(corpus_norm)

Long Time Series

# Show every 5th year on x-axis
curveCtuPlot(
  corpus_norm,
  ctu_noun = c("data", "network", "cloud"),
  r = 5
)

Dark Theme for Presentations

# Use dark theme
curveCtuPlot(
  corpus_norm,
  ctu_noun = c("machine", "learning", "neural"),
  themety = "dark"
)

---

🎯 Selecting Good Example Keywords

Criteria for CTU Selection

1. Zone Distribution - Choose one keyword from high-frequency zone - Choose one from medium-frequency zone - Choose one from low-frequency zone

2. Clear Trajectories - Select keywords with distinct, interpretable patterns - Avoid keywords with flat or noisy trajectories - Choose terms that tell a story

3. Relevance - Select keywords meaningful to your research question - Choose terms your audience will recognize - Pick examples that illustrate interesting phenomena

4. Representativeness - Keywords should be typical of their zone - Avoid extreme outliers within zones - Select terms that generalize well

---

🔍 Interpreting the Plot

Understanding Zone Curves

High-frequency zone (typically darkest): - Shows behavior of core vocabulary - Usually more stable trajectories - Changes indicate major conceptual shifts

Medium-frequency zone: - Shows specialized vocabulary - May show more dynamic patterns - Often contains emerging important terms

Low-frequency zone (typically lightest): - Shows rare or specialized terms - More volatile, less stable - May contain emerging or declining concepts

Understanding Highlighted Keywords

The 3 highlighted keywords provide concrete examples of the abstract zone patterns:

• High-freq example: Demonstrates core vocabulary evolution
• Medium-freq example: Shows specialized term dynamics
• Low-freq example: Illustrates niche or emerging concepts

Look for: - Do example keywords follow their zone’s pattern? - Are there divergences that tell a story? - Do trajectories align with domain knowledge?

---

📈 Use Cases

1. Conference Presentations

Create clear, focused visualizations showing zone structure with examples.

2. Publication Figures

Generate figures demonstrating temporal patterns across frequency strata.

3. Teaching/Demonstrations

Show students or colleagues how frequency zones behave differently.

4. Methodology Validation

Demonstrate that zone classification captures meaningful distinctions.

5. Narrative Building

Tell a story through carefully selected representative keywords.

6. Comparative Analysis

Compare zone patterns across different corpora or time periods.

7. Hypothesis Illustration

Show specific keywords that exemplify theoretical predictions.

---

💡 Tips & Best Practices

1. Choose meaningful examples — Select keywords that resonate with your audience
2. Tell a story — Pick keywords that illustrate your main argument
3. Check zone membership — Verify keywords are in different zones
4. Test readability — Ensure highlighted keywords stand out
5. Match theme to context — Light for papers, dark for presentations
6. Use consistent examples — Keep same keywords across related plots
7. Document selection — Explain why you chose specific keywords
8. Validate patterns — Ensure examples are representative of their zones
9. Save high resolution — Use dpi = 300+ for publications
10. Combine with other plots — Use alongside rowMassPlot() and colMassPlot()

---

📚 See Also

• curvePlot() — Plot all keyword trajectories without zone emphasis
• facetPlot() — Create faceted visualizations by zone
• rowMassPlot() — Visualize keyword frequency distribution
• colMassPlot() — Visualize temporal corpus structure
• normalization() — Normalize frequencies before plotting