engleski

Extracting Domain Knowledge by Complex Networks Analysis of Wikipedia Entries

In this paper we describe a complex networks analysis of Wikipedia. We construct 10 different networks from Wikipedia entries (articles) related to the chosen domain. The goal of the experiment is to extract domain knowledge in terms of identifying entries that are centrally positioned and entries that are strongly related. We apply complex networks analysis on all acquired networks and examine the networks’ structure. We employ centrality measures in order to find centrally positioned entries in the network. Furthermore we identify communities and find which entries are densely connected according to the network structure.

complex networks; domain knowledge extraction; wikipedia

nije evidentirano