A VNS-based quartet algorithm for biomedical literature clustering

CONSOLI, Sergio; STILIANAKIS, Nikolaos

doi:10.1016/j.endm.2014.11.003

This paper proposes a methodology that is able to search for relevant references for systematic reviews and meta-analysis from theMedline/PubMed database, and then to visualize 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. The method allows to retrieve intuitively biomedical-related bibliography, and to detect the structure of the literature collection examined. Keywords: Hierarchical clustering, quartets, variable neighbourhood search, biomedical information extraction, data representation, graphs.

CONSOLI Sergio; STILIANAKIS Nikolaos;

2015-02-24

ELSEVIER BV

JRC91081

1571-0653,

http://www.sciencedirect.com/science/article/pii/S1571065314000456, https://publications.jrc.ec.europa.eu/repository/handle/JRC91081,

10.1016/j.endm.2014.11.003,

Name	Country	City	Type

Datasets

ID	Title	Public URL

Dataset collections

ID	Acronym	Title	Public URL

Scripts / source codes

Description	Public URL

Additional supporting files

File name	Description	File type

Items published in the JRC Publications Repository are protected by copyright, with all rights reserved, unless otherwise indicated. Additional information: https://ec.europa.eu/info/legal-notice_en#copyright-notice