NetworkBeginner

Arc Diagram

A network chart that arranges nodes along a single axis and draws curved arcs between connected pairs — simple, elegant, and easy to read for ordered data.

// 01 — The chart

What it looks like

Example — Character interactions in a novelChapters 1 – 12
AliceBobCarolDaveEveFrank

An arc diagram showing character interactions. Thicker arcs indicate more frequent interactions. Node size reflects total connections.

// 02 — Definition

What is an arc diagram?

An arc diagram is a type of network visualization that places all nodes along a single horizontal (or vertical) line and draws semicircular arcs between connected pairs. The result is a clean, one-dimensional layout that avoids the overlapping clutter of traditional node-link diagrams.

The width or height of each arc corresponds to the distance between the two connected nodes on the line, making it easy to see short-range vs long-range connections at a glance. Arc thickness or color can encode additional information like connection strength or frequency.

Arc diagrams are especially effective when the node ordering is meaningful — for example, along a timeline, a genome sequence, or a narrative structure. They reveal patterns like recurring connections, clusters of tightly linked neighbors, and isolated long-range bridges.

Origin: The arc diagram was introduced by Martin Wattenberg in 2002 in his work “Arc Diagrams: Visualizing Structure in Strings”, originally used to reveal repetitive patterns in music and text sequences.

// 03 — Anatomy

Parts of an arc diagram

ABCDE
A — Nodes: Circles placed along a single axis, each representing an entity in the network
B — Baseline: The horizontal line along which all nodes are positioned in a meaningful order
C — Arcs (edges): Semicircular curves connecting pairs of nodes, encoding relationships in the network
D — Arc thickness: The width of an arc encodes the strength or frequency of the connection
E — Node size: The radius of each node can encode the total number of connections (degree)

// 04 — Usage

When to use it — and when not to

✓Use an arc diagram when…
  • Nodes have a natural linear order (time, sequence, position) that should be preserved
  • You want a compact, clean layout that fits in narrow or horizontal spaces
  • Showing repeating patterns or self-similar structure in sequential data
  • Comparing short-range connections (nearby arcs) vs long-range ones (tall arcs)
  • The network has moderate density — not too sparse, not overwhelmingly dense
  • You want to combine network connections with a parallel timeline or sequence view
×Avoid an arc diagram when…
  • Nodes have no meaningful linear order — the layout will be arbitrary
  • The network is very dense — arcs will stack up and become unreadable
  • You need to show community clusters — force-directed layouts work better
  • The network has more than 50–80 nodes — labels and arcs become crowded
  • You need to show hierarchical relationships — use a tree or dendrogram instead
  • Path-finding or flow direction is the primary analytical goal

// 05 — Reading guide

How to read an arc diagram

Follow these steps whenever you encounter an arc diagram in the wild.

1

Read the node labels along the baseline

Understand what the nodes represent and what order they are in. Is it chronological, sequential, alphabetical? The ordering often carries meaning.

2

Notice arc height and span

Tall arcs connect distant nodes; short arcs connect neighbors. A pattern of mostly short arcs means the network is locally connected. Tall arcs crossing the entire diagram reveal long-range relationships.

3

Check arc thickness or color

Thicker or darker arcs usually indicate stronger connections. Look at the legend to understand the encoding. Some diagrams use color to categorize different types of relationships.

4

Look for clusters of dense arcs

A region with many overlapping arcs indicates a tightly connected group. Gaps between clusters suggest natural community boundaries in the network.

5

Identify isolated nodes

Nodes with no arcs or very few arcs are peripheral to the network. They may be worth investigating — are they disconnected by design or missing data?

// 06 — Common mistakes

Mistakes to watch out for

Arbitrary node ordering

The power of an arc diagram depends entirely on how nodes are ordered. A random or alphabetical order hides patterns. Use a meaningful sequence (time, position, clustering score) or reorder nodes to minimize arc crossings.

Too many arcs without transparency

Dense networks with opaque arcs create an unreadable tangle. Use semi-transparent arcs so overlapping connections remain visible. Alternatively, filter to show only the strongest connections.

Uniform arc styling

If all arcs look identical, the diagram communicates only topology — missing the opportunity to encode edge weight, type, or direction through thickness, color, or opacity.

No node labels

Without labels, nodes are anonymous circles. Even with many nodes, at least label the most connected or most interesting ones. Consider interactive tooltips for the rest.

Ignoring arc direction

In directed networks, arcs should indicate direction (e.g., arcs above the line go left-to-right, below go right-to-left). Failing to show directionality loses critical information about asymmetric relationships.

// 07 — Real-world examples

Where you’ll see arc diagrams used

01

Digital humanities: Narrative structure analysis

Researchers use arc diagrams to show how characters interact across a novel or screenplay. Nodes are characters in order of appearance, and arcs connect characters who share scenes. Clusters of arcs reveal acts or plot threads.

Digital Humanities
02

Genomics: Gene co-expression networks

Bioinformaticians place genes along a chromosome and draw arcs between co-expressed pairs. Short arcs show cis-regulatory relationships (nearby genes) while long arcs highlight trans-regulatory connections spanning the genome.

Bioinformatics
03

Music analysis: Repetitive structure in compositions

Martin Wattenberg's original application showed repeated musical phrases as arcs connecting matching passages. The resulting pattern instantly reveals the verse-chorus structure and thematic repetitions in a piece of music.

Music & Art

// 08 — At a glance

Quick reference

Also known asLinear network diagram, arc plot
Invented byMartin Wattenberg, 2002
Best forNetworks with a meaningful linear node order
Data typesNetwork/graph data with ordered nodes and edges
Recommended nodes10 – 60 nodes for readability
Arc height encodesDistance between connected nodes along the axis
Common toolsD3.js, Observable, R (ggraph), Python (networkx + matplotlib)
Common mistakesRandom ordering, opaque dense arcs, no labels, uniform styling

// 09 — Variations

Types of arc diagrams

The basic arc diagram has several useful variants adapted for different data types and analytical goals.

Weighted arc diagram

Arc thickness encodes edge weight, letting you compare strong vs weak connections at a glance.

Double-sided arc diagram

Arcs above and below the line distinguish directed edges or two categories of relationships.

Vertical arc diagram

Nodes run top-to-bottom instead of left-to-right, useful for sequences that read vertically like timelines.

Sized-node arc diagram

Node radius encodes a second variable like degree centrality or total weight, adding a layer of information.

// 10 — FAQs

Frequently asked questions

What is an arc diagram?+

An arc diagram is a type of network visualization that places all nodes along a single horizontal (or vertical) line and draws semicircular arcs between connected pairs. The result is a clean, one-dimensional layout that avoids the overlapping clutter of traditional node-link diagrams.

When should you use an arc diagram?+

Use an arc diagram when nodes have a natural linear order (time, sequence, position) that should be preserved. It also works well when you want a compact, clean layout that fits in narrow or horizontal spaces, and when showing repeating patterns or self-similar structure in sequential data.

When should you avoid an arc diagram?+

Avoid an arc diagram when nodes have no meaningful linear order — the layout will be arbitrary. It is also a poor fit when the network is very dense — arcs will stack up and become unreadable, or when you need to show community clusters — force-directed layouts work better.

Is an arc diagram suitable for dashboards?+

Yes — an arc diagram can work well in dashboards as long as the panel is large enough for readers to perceive the encoded values, has a clear title, and includes the legend or axis labels needed to interpret it.

What category of chart is an arc diagram?+

Arc Diagram belongs to the Network family of charts. Charts in that family are designed to answer the same kind of question, so they often work as alternatives when one doesn't quite fit your data.

How do you read an arc diagram?+

Start with the axis labels and legend, then look at the overall shape before zooming into individual marks. Compare prominent features against the rest of the data, and verify any conclusion against the underlying numbers when precision matters.