A VNS-based quartet algorithm for biomedical literature clustering

This paper proposes a methodology that is able to search for relevant references for systematic reviews and meta-analysis from Medline/PubMed, and then to represent the retrieved bibliography through the quartet method of hierarchical clustering. As this novel approach is based on a NP-hard combinatorial problem, a Reduced Variable Neighbourhood Search is used to produce the graph of document clusters as output from the input distance matrix whereby the number of clusters is not known in advance. The distance matrix is derived from the link-ranking XML data returned by PubMed with the search results.