In a Wikipedia link graph, each node is an article page. If article A links to article B, the graph contains a directed edge from A to B. The reverse edge exists only if B also links back to A.

This model is useful for exploring navigation structure. It does not by itself prove that two ideas have a particular semantic relationship; it records a link between pages.

A sampled graph is not the whole encyclopedia

Wikipedia contains far more articles and links than a typical interactive view should load at once. A crawler therefore chooses a starting page, follows links to a limited depth, and applies a page limit.

Metrics such as PageRank, betweenness, or community assignments describe the sampled subgraph being analyzed. A different starting article or limit can produce a different structure.

Explore a topic map

WikiCrawl can map articles around quantum mechanics, climate change, or astronomy. The interactive graph preserves link direction and lets you inspect pages and their positions in the current sample.