yfiles.com

Graph visualization is the process of drawing a graph — a set of vertices (nodes) connected by edges — as a diagram so you can see its structure and reason about it. You use it when you need to inspect relationships (routes, dependencies, networks) rather than read them from a table. On a tool like Graph Online (graphonline.top), you can build that graph in the browser, run algorithms such as shortest path, and export the result as an image or file. This article covers what the tool accepts as input, how to run shortest path, and the format errors that most often block a first attempt.

What graph visualization is used for

A graph is an abstract structure: vertices plus the edges between them. Visualization turns that structure into something you can look at, which matters because many graph properties are hard to spot in raw data but obvious in a picture.

Typical uses:

• Path and routing problems — find the shortest path between two vertices, e.g. a route between two locations.
• Connectivity checks — search connected components to see whether the graph splits into isolated groups.
• Structural summaries — compute an adjacency matrix, incidence matrix, or distance matrix to feed other tools.
• Optimization — find a minimum spanning tree to connect all vertices at minimum total edge weight.
• Traversal problems — look for an Eulerian cycle, which visits every edge exactly once.

The point of doing this online is that you skip installing and configuring graph software: you enter the graph, run the algorithm, and read the result in the same workspace.

Ways to enter a graph

Graph Online accepts several input methods, listed on the site as: using an adjacency matrix, using an incidence matrix, using an edge list, or starting from a graph example. You can also draw the graph directly in the workspace with Add vertex and Connect vertices.

For weighted graphs, the adjacency matrix holds edge weights; the site notes that a multigraph matrix contains the weight of the minimum edge between vertices.

Creating a graph and finding the shortest path

The site's heading for this workflow is Find shortest path, and the Help page links to a tutorial video. The general steps:

1. Choose an input method — adjacency matrix, incidence matrix, edge list, or draw manually.
2. Enter the data. For matrix input, use a comma as the separator. The matrix must be square (for the adjacency matrix).
3. Confirm the graph is created. If the format is wrong, the tool reports it rather than drawing a partial graph (see the error section below).
4. Select the algorithm. Open the Algorithms menu and choose shortest path.
5. Read the result in the workspace — the path is highlighted on the visualization.
6. Save or export if you need the result outside the browser.

Saving and exporting the result

The workspace offers several output options:

• Save graph image — a printed image of the graph.
• Save graph to SVG — vector output, useful if you need to scale or edit it.
• Workspace screenshot — captures the current view.
• Export to file / Import from file — move a graph between sessions.
• Save — the site confirms with a message and a link to view the saved graph, and the image is available at that link.

There is also a Paste your algorithm box: you paste an algorithm and press Run. Submitted algorithms are sent for checking and, if successful, added to the site.

Common input errors and how to fix them

Most first failures are format problems, not algorithm problems. The tool reports each one specifically:

• Adjacency matrix has wrong format — click fix matrix to attempt a fix, or open Help for the expected format.
• Incidence matrix has wrong format — same options: fix matrix or Help.
• Edge list has wrong format — click fix or open Help for the format.
• Matrix is incorrect — check that you used a comma as the separator.
• Matrix should be square — the adjacency matrix must have the same number of rows and columns.
• The graph is too large for this algorithm — the tool states the vertex and edge counts and the maximum size the algorithm supports. Reduce the graph or pick a different algorithm.

A practical rule: if the graph does not appear at all, the problem is almost always the input format, so check the separator and the square-matrix requirement before changing anything else.

Practical notes

• The workspace includes view controls (zoom in/out, zoom fit, 100%/50%/25%) and style options (Default, Night, Large, High Contrast), which help when a graph is dense or you need a readable export.
• Undo is available with Ctrl+Z.
• The site is open source and has a Wiki and News section, so you can check for changes or contribute.
• If something breaks, the page provides a contact address for reporting problems.

If your goal is simply to see a graph and get a shortest path, the fastest route is: enter an edge list or adjacency matrix, confirm the graph renders, then run shortest path from the Algorithms menu. If the graph fails to render, fix the format first — the error message tells you which of the three input types is malformed.