Please use this identifier to cite or link to this item:
Title: Document clustering via multiple correspondence, term and metadata analysis in R
Authors: Koutsoupias, Nikos 
Mikelis, Kyriakos 
Author Department Affiliations: Department of International & European Studies 
Department of International & European Studies 
Author School Affiliations: School of Social Sciences, Humanities and Arts 
School of Business Administration 
Subjects: FRASCATI__Natural sciences__Computer and information sciences
Keywords: document clustering
hierarchical clustering
multiple correspondence analysis
document metadata
text mining
Issue Date: 2019
Conference: 16th Conference of the International Federation of Classification Societies 
We introduce the combined use of multiple correspondence analysis, metadata and term frequencies for clustering articles of a scientific journal. A period of five years (2010-2014) is covered, with approximately 125 articles. Through specific R packages for multidimensional data analysis and text mining, the approach links quantitative analysis of discourse to clustering documents considering both metadata and frequent terms.
DOI: 10.13140/RG.2.2.22716.59527
Rights: Attribution-NonCommercial-NoDerivatives 4.0 International
Corresponding Item Departments: Department of International & European Studies
Department of International & European Studies
Appears in Collections:Conference proceedings

Files in This Item:
File Description SizeFormat
[Σ38].pdf1,44 MBAdobe PDF
Show full item record

Page view(s)

checked on Jul 25, 2024


checked on Jul 25, 2024

Google ScholarTM




This item is licensed under a Creative Commons License Creative Commons