What you will see is a graphical representation of Wikipedia. Each circle represents an article in the English Wikipedia.
The size of a circle depends on how many real pageviews it got on a particular day. For example you will see that the Main Wikipedia article is the biggest one as it gets the most views.
The placement of the circles is based on the links between the articles. The more closely linked the two articles are, the closer they appear.
The current algorithm for the placement of the circles is naive, includes randomness and is not totally accurate. This will be improved (hopefully).
Warnings
Mobile is supported, but I highly recommend using a laptop/desktop for the best experience.
Due to the nature of Wikipedia, the dataset contains NSFW elements.
The placement of the articles is not perfect; you might see some articles wildly out of place.
The dataset is from September 10, 2024 and is not updated, so the articles might differ from current Wikipedia.
The dataset may contain errors.
The placement of nodes is only a result of the algorithm; it is not manually adjusted.
An example of this is the Twitter article. Although you would expect it to be near other social media apps, it is near anime communities. The reason for this is that many anime Wikipedia pages have links to twitter in their reference section. Many anime news come from twitter which means that nearly every anime article has a link to the twitter article on Wikipedia. This makes twitter appear very close on anime communities. There are many fixes for this but I am yet to implement a solution. This should be improved on future updates.
How to Use
Desktop: Use your mouse to pan around and your mouse wheel to zoom in and out.
Laptop: Use your touchpad to pan around and two fingers to zoom in and out.
Mobile: Use a finger to pan around and pinch to zoom.
Search specific articles using the Search bar.
Click any circle to open the Wikipedia article.
If the visualization runs slowly, try a smaller dataset.
Technical Details
Data is collected from Wikimedia dumps
Uses Leiden algorithm for community detection (and to build the graph itself)