Link prediction in co-authoring networks: a comparative analysis using two versions of topological metrics

Authors

  • Mariana Magalhães de Mattos Coelho Instituto Militar de Engenharia https://orcid.org/0000-0003-1838-3145
  • Claudia Marcela Justel Instituto Militar de Engenharia

DOI:

https://doi.org/10.5335/rbca.v15i2.13817

Keywords:

Social Network Analysis, Graph Applications, Topological Metrics, Link Prediction

Abstract

The problem called link prediction consists of estimating the appearance of of edges between nodes of a graph representing a network. Among the different approaches of the problem proposed in the literature, we consider only the topological one. We use topological metrics in two versions: traditional and in pairs, (the last version in two variants, ‘or’ and ‘and’). The objective of this work is to compare four local topological metrics, in two versions, performing experiments in five real co-authorship networks. We present the results and conclusions obtained from the experiments performed on five real ArXiv networks.

Downloads

Download data is not yet available.

Published

2023-07-27

Issue

Section

Original Paper

How to Cite

[1]
2023. Link prediction in co-authoring networks: a comparative analysis using two versions of topological metrics. Brazilian Journal of Applied Computing. 15, 2 (Jul. 2023), 11–21. DOI:https://doi.org/10.5335/rbca.v15i2.13817.