An Improved Algorithm for Extracting Research Communities from Bibliographic Data

摘要

In this paper we improve the performance of the community extraction algorithm in [1] from bibliographic data, which was originally proposed for web community discovery by [2]. A web community is considered to be a set of web pages holding a common topic, in other words, it is a dense subgraph induced in web graph. Such subgraphs obtained by the max-flow algorithm are called max-flow communities, and this algorithm was improved to obtain research communities from bibliographic data by the strategy for selection of community nodes in [1]. We propose an improvement of this algorithm by carefully selecting initial seed node, and show the performance of this algorithm by experiments for the list of many keywords frequently appearing in data.