wiki_topcats — Wikipedia hyperlinks among pages of top categories (2011)
- Description
A directed network of hyperlinks between articles of the English Wikipedia, collected in September 2011 and distributed by the Stanford Network Analysis Project (SNAP) as
wiki-topcats, as a benchmark for clustering with ground-truth communities. The network was constructed by taking the largest strongly connected component of the hyperlink network of Wikipedia, restricting it to the articles in the top categories (those with at least 100 articles), and taking the largest strongly connected component of the result. Nodes are articles, and an edge $i \to j$ means that article $i$ links to article $j$. The categories of the articles (which overlap, since an article can belong to several categories) are given as ground-truth communities; only the top categories are included.The network has the following properties:
- (Node)
name: the title of the article. - (Node)
categories: the top categories of the article (e.g. "Category:Skin_conditions_resulting_from_physical_factors").
- (Node)
- Tags
- Informational Web graph Unweighted Metadata
- Added
- 2026-10-27
- Citation
-
- H. Yin, A. R. Benson, J. Leskovec, and D. F. Gleich, "Local Higher-Order Graph Clustering." Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 555-564 (2017). 10.1145/3097983.3098069 Sci-Hub
- C. Klymko, D. F. Gleich, and T. G. Kolda, "Using Triangles to Improve Community Detection in Directed Networks." Proceedings of the ASE BigData Conference (2014). arXiv:1404.5874
- Upstream URL OK
- https://snap.stanford.edu/data/wiki-topcats.html
- Networks
-
Tip: hover over a column header for a description.
Column legend
- $\left<k\right>$
- Mean degree
- $\sigma_k$
- Standard deviation of the degrees
- $\lambda_h$
- Largest eigenvalue of the Hashimoto (non-backtracking) matrix
- $\tau$
- Random walk mixing time
- $r$
- Degree assortativity
- $c$
- Global clustering coefficient
- $\oslash$
- (Pseudo-) diameter
- $S$
- Relative size of the largest component
- NPs
- Node properties
- EPs
- Edge properties