Networks / Pajek
Package for Large Network Analysis


  1. Krebs1.pdf, Krebs2.pdf
    Internet industries (collected by Valdis Krebs) 219 vertices, 631 edges. Each node in the network represents a company that competes in the Internet industry, 1998 to 2001; red - content, blue - infrastructure, yellow - commerce. Two companies are connected with an edge if they have announced a joint venture, strategic alliance or other partnership.. The vertex size is proportional to its betweenness. Krebs2.pdf contains a zoom view of the main part.
  2. eatWork.pdf, eatMoney.pdf, eatCar.pdf
    Selected thematic islands from EAT (The Edinburgh Associative Thesaurus) network: 23219 vertices, 325624 arcs, arc weights are the response counts. More.
  3. ODLIS.pdf
    Line cut at level 11 of transitive triangular network derived from ODLIS (Online Dictionary of Library and Information Science): 2909 vertices, 18419 arcs. More.
  4. Reuters.pdf
    500-core of the Terror news network (66 days combined; 13332 vertices, 243447 edges). The weight of edges is the frequency of co-appearance of given two words linked by the edge. The t-core is the maximum subnetwork in which each its vertex co-appeared in the text at least t times with other vertices from the subnetwork.
  5. SOM.pdf
    Main subnetwork at level 0.007 of the SOM (selforganizing maps) citation network (4470 vertices, 12731 arcs). The arc weights are proportional to the number of different source-sink paths passing through the arc. More.
  6. CompGeom.pdf
    pS-core at level 46 of Geombib (Computational Geometry) network ( 7343 vertices, 11898 edges, edge weight counts the number of common works). More.

How to?; Pajek; Vlado/Networks